# Proxoria — full connector reference Proxoria connects AI assistants (Claude, ChatGPT, or any Model Context Protocol client) to the business apps you already use. Each connector turns a third-party API into tools your AI can call, with per-user authentication and write actions gated by human approval. No official third-party MCP server is required — Proxoria provides the connector. --- ## Aircall URL: https://www.proxoria.com/en/connectors/aircall Connect your AI to Aircall. Aircall cloud phone: call logs with their recordings, transcriptions and AI analysis (summary, topics, sentiment, action items), contacts, numbers, tags, teams and user availability. Authentication: connect Aircall with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_calls (read-only): List calls (paginated). Optionally filter by a Unix-time range. - get_call (read-only): Get a single call, including its recording and transcription when available. - list_contacts (read-only): List contacts (paginated). - list_users (read-only): List the users (agents) of the Aircall account. - add_call_comment (write, needs approval): Add a comment to a call (e.g. an AI-written summary). Writes to Aircall, so it goes through the confirmation guardrail when enabled. - get_analytics_export (read-only): Fetch a previously requested analytics report export. Exports are asynchronous: this returns the file once it is ready. - search_calls (read-only): Search calls by direction, status, user, number, tag or time range. - get_call_action_items (read-only): Get the action items Aircall extracted from a call. - get_call_custom_summary (read-only): Get the result of a custom summary template applied to a call. - get_call_evaluations (read-only): Get the quality evaluations filled in on a call by a reviewer. - get_call_playbook_result (read-only): Get how a call scored against a conversation playbook — which expected steps the agent covered. - get_call_predicted_csat (read-only): Get the satisfaction score Aircall predicts for a call, without asking the customer. - get_call_realtime_transcription (read-only): Get the live transcription captured while the call was running, if real-time transcription is on. - get_call_sentiments (read-only): Get the sentiment analysis of a call. - get_call_summary (read-only): Get Aircall's automatic summary of a call. - get_call_topics (read-only): Get the topics Aircall detected in a call. - get_call_transcription (read-only): Get the transcription of a call — the text of what was said, speaker by speaker. - tag_call (write, needs approval): Set the tags of a call. THIS REPLACES THE WHOLE SET: pass every tag id the call should keep, not only the new ones. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - create_contact (write, needs approval): Create a contact. Contacts created through the API are automatically shared across the whole company. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - search_contacts (read-only): Search contacts by name, email or phone number. - get_contact (read-only): Get one contact. Only contacts shared across the company are reachable. - update_contact (write, needs approval): Update a contact's fields. Aircall uses POST, not PATCH, for this update. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - add_contact_email (write, needs approval): Add one email address to a contact. A contact holds at most 20; beyond that the call is rejected. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - delete_contact_email (write, needs approval): Remove one email address from a contact, by the email entry's own id. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - update_contact_email (write, needs approval): Update one email address of a contact. The `value` must be sent every time, even when only the label changes. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - add_contact_phone (write, needs approval): Add one phone number to a contact. A contact holds at most 20; beyond that the call is rejected. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - delete_contact_phone (write, needs approval): Remove one phone number from a contact, by the phone entry's own id. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - update_contact_phone (write, needs approval): Update one phone number of a contact. Aircall normalises the value before storing it, and it must be sent every time. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - list_numbers (read-only): List the numbers of the account with their configuration. - get_number (read-only): Get one number and its configuration. - update_number_messages (write, needs approval): Update the music and messages of a number (greeting, waiting music, voicemail). Writes to Aircall, so it goes through the confirmation guardrail when enabled. - get_number_messaging_config (read-only): Read the messaging configuration of a number (SMS and WhatsApp capabilities). - get_number_registration_status (read-only): Check where a number stands in the regulatory registration process (10DLC and equivalents). - list_whatsapp_templates (read-only): List the WhatsApp message templates approved for a number. - get_whatsapp_status (read-only): Check whether the WhatsApp line attached to a number is active. - list_sms_templates (read-only): List the account's SMS templates. - list_tags (read-only): List the account's tags with their ids — needed to tag a call, which takes ids and not names. - create_tag (write, needs approval): Create a tag. Aircall validates it before inserting: a name is required, and the colour must be hexadecimal. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - delete_tag (write, needs approval): Delete a tag permanently. It disappears from every call it was applied to. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - get_tag (read-only): Get one tag. - update_tag (write, needs approval): Rename a tag or change its colour. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - list_teams (read-only): List the teams of the account and their members. - remove_user_from_team (write, needs approval): Remove a user from a team. The user keeps their account; only the team membership goes. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - add_user_to_team (write, needs approval): Add a user to a team. Writes to Aircall, so it goes through the confirmation guardrail when enabled. - list_user_availabilities (read-only): List the availability of every user — who can take a call right now. - get_user_availability (read-only): Check whether one user is available to take a call right now. - get_dialer_campaign (read-only): Get the dialer campaign currently loaded for a user. - list_dialer_campaign_numbers (read-only): List the phone numbers queued in a user's dialer campaign. - get_user (read-only): Get one user, on the v2 route — v1 users is deprecated. Example prompts: - "List calls (paginated). Optionally filter by a Unix-time range — from Aircall." - "Get a single call, including its recording and transcription when available — from Aircall." - "List contacts (paginated) — from Aircall." --- ## BambooHR URL: https://www.proxoria.com/en/connectors/bamboohr Connect your AI to BambooHR. BambooHR: the employee directory and the tables behind each profile, time off with balances and holidays, time tracking down to clock entries and breaks, scheduling, goals and training, compensation — planning cycles, pay bands, benchmarks, benefits and total rewards, read-only — saved reports and queryable datasets, the hiring pipeline, and the field vocabulary that makes the rest legible. Authentication: connect BambooHR with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_employees (read-only): List the employee directory (all employees with their basic fields). - get_employee (read-only): Get one employee's details. Pass a comma-separated `fields` list to customize. - list_time_off_requests (read-only): List time-off requests, optionally filtered by date range and status. - who_is_out (read-only): Who is out (time off and holidays) over a date range. - list_all_employees (read-only): List every employee, page by page, with more than the published directory shows. Use list_employees for the quick directory; this one when the directory's fields are not enough or when someone is missing from it. - update_employee (write, needs approval): Change an employee's fields. The names come from list_fields — BambooHR takes them verbatim and silently ignores one it does not know, so check the spelling there first. What a key may change is limited by its own permissions: pay and other sensitive fields are refused unless the key was granted them. - get_employee_table (read-only): Read one of an employee's tables — the rows behind a tab of their profile: job history, compensation, education, emergency contacts, assets. `list_tabular_fields` names the tables that exist and what each holds. - list_changed_employees (read-only): List the employees whose record changed since a moment — the way to catch up without re-reading everyone. Returns ids and what happened to each: inserted, updated, deleted. - list_changed_table_data (read-only): List the rows of one employee table that changed since a moment — for instance every compensation change since the start of the year, across the company. - get_time_off_balance (read-only): Get someone's remaining time-off balance, by policy, as it stands on a date — the answer to 'how many days do I have left'. list_time_off_requests shows what was asked for; this shows what is left. - list_time_off_types (read-only): List the kinds of time off the company has defined — holiday, sick, unpaid. - list_time_off_policies (read-only): List the time-off policies — how each kind of leave accrues, which is what makes a balance mean something. - list_reports (read-only): List the saved custom reports. Running one is how you answer a question with the company's own definitions rather than rebuilding them. - run_report (read-only): Run a saved report and get its rows. - list_datasets (read-only): List the datasets that can be queried — BambooHR's own analytical views over employee, time-off and payroll data. - get_dataset_fields (read-only): List the fields of a dataset, with their types. Read this before query_dataset: the filters and the columns you ask for must name fields that exist here. - query_dataset (read-only): Query a dataset: choose the fields to return, filter on them, and group or sort. This is the closest thing to asking BambooHR a question of your own rather than reading a report someone else defined. - get_dataset_field_options (read-only): List the values a dataset field can take — the allowed entries of a dropdown. Ask for these before filtering on such a field, so the filter matches something. - list_job_openings (read-only): List the job openings, with their status, location and hiring lead. - create_job_opening (write, needs approval): Open a new position for applications. The hiring lead comes from list_hiring_leads and the location from list_hiring_locations. BambooHR takes this one as multipart form data, not JSON, because the same endpoint can carry files; no file is sent here. - list_applications (read-only): List job applications, with the candidate, the job and the current status. - get_application (read-only): Get one application in full, with the candidate's details and their answers. - add_application_comment (write, needs approval): Add a comment to an application — the running commentary a hiring team leaves on a candidate. Visible to everyone with access to the hire. - update_application_status (write, needs approval): Move an application to another status — advancing a candidate, or turning them down. The status ids come from list_applicant_statuses. Rejecting may trigger whatever notification the company has configured, so this is a decision, not a filing action. - list_applicant_statuses (read-only): List the statuses an application can be in, with their ids. - list_hiring_locations (read-only): List the company locations a job opening can be attached to. - list_hiring_leads (read-only): List the employees who can be named as the lead on a hire. - create_candidate (write, needs approval): Add a candidate and their application to a job opening — someone who applied elsewhere, or a referral being entered by hand. BambooHR takes this one as multipart form data, not JSON, because the same endpoint can carry a résumé; no file is sent here, so attach the CV from BambooHR afterwards if there is one. - list_employee_files (read-only): List an employee's documents and the categories they are filed under — contracts, certificates, reviews. Returns what each file is, not its content. - list_company_files (read-only): List the company-wide documents and their categories. - list_fields (read-only): List every employee field the company has, with its name and type. This is the vocabulary: the names update_employee and the `fields` parameters expect. - list_tabular_fields (read-only): List the employee tables and the fields inside each — which tables get_employee_table can read, and what they hold. - list_list_fields (read-only): List the fields that are dropdowns, with the values each one allows. Read this before writing to such a field: BambooHR accepts only the exact values defined here. - list_users (read-only): List the BambooHR users of the account, with their access level and status. - list_bank_holidays (read-only): List the public holidays BambooHR knows for a year — what makes a working day different from a calendar day when counting leave. The country comes from the account, it is not a parameter. - get_countries (read-only): Returns a JSON array of every country in the catalog, or a single country object when `isoCode` is supplied. - get_country_by_id (read-only): Returns a single country from the catalog as a JSON object, identified by its numeric id. - list_company_benefits (read-only): Returns all active (non-deleted) company benefit plans for the account. - list_employee_benefits (read-only): Returns current and scheduled-future benefit enrollment records, grouped by employee. - list_member_benefit_events (read-only): Returns benefit enrollment events for all employees and their dependents over the past year, organized by member. - list_benefit_coverages (read-only): Returns all benefit coverage levels configured in the company, such as Employee Only, Employee + Spouse, and Employee + Family. - list_member_benefits (read-only): Returns a paginated list of benefit enrollment records for all members (employees and dependents) in the company for a given calendar year. - list_benefit_deduction_types (read-only): Returns all benefit deduction types available in the system. - list_employee_dependents (read-only): Returns employee dependents for the company. - get_employee_dependent (read-only): Returns the details of a single employee dependent by their dependent ID. - list_calendar_events (read-only): Lists calendar events (time off, holidays, birthdays, and anniversaries) overlapping the requested date range. - get_company_equity_settings (read-only): Retrieves company-level equity settings including calculation type, valuation, shares, pricing, and vesting-related configuration. - list_available_compensation_tools (read-only): Returns the list of available compensation tools/settings for the company, including Levels & Bands, Compensation Benchmarking, Compensation Planning, and Total Rewards. - list_compensation_benchmarks (read-only): Returns every job/location pair the company tracks for compensation benchmarking. - get_compensation_benchmark_details (read-only): Returns detailed benchmark data for a single company job title, optionally scoped to a specific location. - export_compensation_benchmark_details (read-only): Returns a CSV export of the compensation benchmark detail view for a single job, optionally scoped to a specific location. - list_compensation_benchmark_sources (read-only): Returns every enabled benchmark source configured for the company. - list_compensation_planning_cycles (read-only): List all compensation planning cycles for the company, including status and employee counts. - get_compensation_planning_cycle_details (read-only): Get compensation planning cycle details and currency settings for a cycle. - list_compensation_planning_cycle_admins (read-only): List compensation planning cycle admins. - get_compensation_planning_approval_flows (read-only): Get approval flows for the compensation planning cycle (grouped by department by default). - get_compensation_planning_cycle_budgets (read-only): Get budget guidelines and breakdown for the compensation planning cycle. - get_change_communication_letter_details (read-only): Get change communication letter details for the compensation planning cycle. - list_employees_in_compensation_planning_cycle (read-only): List employees in the compensation planning cycle (picker data: company roster and membership). - get_compensation_planning_cycle_summary (read-only): Get compensation planning cycle summary (progress, stats). - get_compensation_planning_cycle_worksheet (read-only): Get compensation planning worksheet details for the cycle (recommendations, approvals, or overview context). - export_compensation_planning_cycle_worksheet_to_csv (read-only): Download compensation planning worksheet data as a UTF-8 CSV (Excel-friendly BOM). - list_custom_fields (read-only): Returns a paginated list of active custom fields. - list_custom_field_types (read-only): Returns an object containing the custom field types available when creating a custom field. - get_custom_field (read-only): Returns one custom field definition. - list_custom_field_list_values (read-only): Returns the dropdown options (and employee counts) for a list-type custom field. - get_legacy_report_field_map (read-only): Returns the mapping from the field identifiers the deprecated Reports > Get Company Report (`get-company-report`) endpoint emitted to the field names the Custom Reports and Datasets endpoints use today. - get_legacy_report_id_map (read-only): Returns a mapping from legacy custom report IDs to the new report IDs created by the report migration. - get_report_by_id (read-only): Executes a saved custom report and returns its data using the report's configured fields and filters. - list_custom_tables (read-only): Returns a paginated list of active custom tables with their fields. - get_custom_table (read-only): Returns one custom table definition with its fields. - list_goals (read-only): Returns goals for an employee. - create_goal (write, needs approval): Create a new goal for an employee. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_alignable_goal_options (read-only): Returns goals that can be used as alignment targets. - get_goal_creation_permission (read-only): Determine if the authenticated caller has permission to create a goal for this employee. - list_goal_sharing_options (read-only): Provides a list of employees with whom the specified employee\'s goals may be shared. - delete_goal (write, needs approval): Permanently deletes a goal for an employee. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_goal_aggregate (read-only): Returns a single goal with its comments, alignment options, and a list of all persons who are either shared on the goal or have commented on it. - close_goal (write, needs approval): Closes a goal, moving it to the closed status. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_goal_comments (read-only): Returns comments for a goal, including each comment's ID, author, text, creation time, and edit/delete permissions. - create_goal_comment (write, needs approval): Creates a new comment on a goal. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_goal_comment (write, needs approval): Deletes a goal comment. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - update_goal_comment (write, needs approval): Updates the text of an existing goal comment. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - update_milestone_progress (write, needs approval): Update the progress of a milestone in a goal. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - update_goal_progress (write, needs approval): Update the progress percentage of a simple, non-milestone goal. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - reopen_goal (write, needs approval): Reopens a closed goal, returning it to the in-progress status. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - update_goal_sharing (write, needs approval): Replaces the full list of employees this goal is shared with. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - update_goal (write, needs approval): Update a goal's top-level fields and optionally add or delete milestones. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_goals_aggregate (read-only): Provides a goals dashboard for an employee: goals matching the selected filter, status filter counts/actions, comment counts, and people shared with or commenting on those goals. - get_goal_filters (read-only): Get the number of goals per status for an employee, including goals with milestones. - list_company_holidays (read-only): Returns a paginated list of active company holidays. - create_company_holiday (write, needs approval): Creates a company holiday. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - bulk_create_company_holidays (write, needs approval): Creates multiple company holidays in one synchronous call. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_catalog_holidays (read-only): Lists entries in the global holiday catalog. - get_catalog_holiday (read-only): Gets a global holiday catalog entry by UUID. - delete_company_holiday (write, needs approval): Soft-deletes a company holiday and removes its audience, pay, and country sub-rows. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_company_holiday (read-only): Gets a company holiday by ID. - update_company_holiday (write, needs approval): Updates a company holiday with a JSON Merge Patch (RFC 7396) document. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - create_hour_record (write, needs approval): Adds a single hour record. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - update_hour_record (write, needs approval): Edits an existing hour record by `timeTrackingId`. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_hour_record (write, needs approval): Deletes an hour record by `timeTrackingId` (`id` path parameter). Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - create_or_update_hour_records (write, needs approval): Bulk add/edit hour records. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_time_tracking_record (read-only): Retrieves a single time tracking hour record by its ID. - list_job_locations (read-only): Retrieves a paginated list of job locations. - get_a_job_location (read-only): Retrieves a single job location by its ID. - list_break_assessments (read-only): Returns a paginated list of break assessments. - list_break_policies (read-only): Returns a paginated list of all break policies. - create_break_policy (write, needs approval): Create a break policy. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_break_policy (write, needs approval): Deletes a break policy by its UUID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_break_policy (read-only): Retrieves a single break policy by its UUID. - update_break_policy (write, needs approval): Partially updates a break policy identified by its UUID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - assign_employees_to_break_policy (write, needs approval): Assigns employees to a break policy. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - set_employees_for_break_policy (write, needs approval): Sets the employee assignments for a break policy. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_breaks_for_break_policy (read-only): Returns a paginated list of breaks belonging to the specified break policy. - create_break (write, needs approval): Creates a new break and associates it with the specified break policy. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - replace_breaks_for_break_policy (write, needs approval): Replace all breaks for a break policy. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_break_policy_employees (read-only): Retrieves employees assigned to a specific break policy. - sync_break_policy (write, needs approval): Performs a full replacement of a break policy and its related data (breaks and employee assignments). Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - unassign_employees_from_break_policy (write, needs approval): Unassigns the specified employees from a break policy. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_break (write, needs approval): Deletes a time tracking break by its UUID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_break (read-only): Retrieves a single time tracking break by its UUID. - update_break (write, needs approval): Partially updates a time tracking break identified by its UUID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_employee_break_availabilities (read-only): Retrieves break availability information for an employee. - list_employee_break_policies (read-only): Retrieves break policies assigned to a specific employee. - get_published_levels_and_bands (read-only): Returns the currently published pay grades and bands as a JSON object with a `groups` array. - get_job_titles_and_level_assignments (read-only): Returns the working draft configuration showing which job titles are assigned to each compensation level, as used by the pay grades and bands setup wizard. - list_job_titles_with_employees (read-only): Returns every active, non-archived company job title (not only titles used in the pay grades and bands configuration) together with the employees who currently hold each title. - list_compensation_level_groups_and_levels (read-only): Returns the working draft configuration of compensation level groups and levels, including the per-group and per-level validation `errors` and `warnings` used by the setup wizard. - get_pay_bands (read-only): Returns the working draft pay band configuration for every compensation level, including the pay-band-step validation `errors` and `warnings` used by the setup wizard. - get_levels_and_bands_review (read-only): Returns the pre-publish review of compensation level groups and levels with validation `errors` and `warnings` consolidated across every setup-wizard step: level naming, group-level checks, and pay band values. - get_levels_and_bands_status (read-only): Returns the configuration status of the Pay Grades & Bands (levels and bands) setup, broken down by setup step. - get_compensation_level_group_status_counts (read-only): Returns the number of compensation level groups in each lifecycle status. - list_schedules (read-only): Retrieves a paginated list of all schedules for the company. - create_schedule (write, needs approval): Creates a new schedule for the company. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_schedule (write, needs approval): Deletes a schedule by its UUID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_schedule (read-only): Retrieves a specific schedule by its UUID. - update_schedule (write, needs approval): Partially updates an existing schedule by its UUID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_schedule_pdf (read-only): Generates and streams a PDF of the schedule view for the given schedule and date range. - list_shift_assessments (read-only): Lists shift assessments matching the given filters. - list_shifts (read-only): Lists shifts matching the given filters. - create_shift (write, needs approval): Creates a new shift in the specified schedule. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - publish_shifts (write, needs approval): Publishes one or more planned shifts, making them visible to employees. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_shift (write, needs approval): Deletes a shift by its UUID or composite recurrence ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_shift (read-only): Retrieves a single shift by its UUID. - update_shift (write, needs approval): Updates a shift by its UUID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_timezones (read-only): Returns a paginated list of timezones available for use when creating or editing schedules. - get_break_policy_suggestions (write, needs approval): Uses an AI agent to analyze existing break policies and company context, then returns structured meal and rest break policy recommendations ready for form pre-fill. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_clock_entries (read-only): Returns a paginated list of time tracking clock entries. - create_clock_entry (write, needs approval): Manually creates a time tracking clock entry (corrections, retroactive entry). Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_clock_entry (write, needs approval): Deletes a time tracking clock entry by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_clock_entry (read-only): Retrieves a single clock entry by its ID. - update_clock_entry (write, needs approval): Partially updates a time tracking clock entry identified by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - clock_in (write, needs approval): Clocks an employee in at the current server time, creating an open clock entry (`end: null`). Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - clock_out (write, needs approval): Clocks an employee out at the current server time, closing the employee's open clock entry (`end` set, `endSource: USER`). Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_configurations (read-only): Returns a paginated list of time tracking configurations. - create_configuration (write, needs approval): Creates a new GROUP time tracking configuration together with its approval workflow. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_configuration (write, needs approval): Soft-deletes an empty GROUP time tracking configuration. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_configuration (read-only): Retrieves a single time tracking configuration by its ID. - update_configuration (write, needs approval): Updates a time tracking configuration using JSON Merge Patch (RFC 7396) semantics: only the properties present in the body are applied, omitted properties are unchanged, and an explicit null clears a nullable property. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_enrolled_employees (read-only): Returns a paginated list of employee time tracking enrollments. - bulk_upsert_employee_enrollments (write, needs approval): Bulk enables, disables, or reassigns employee enrollments. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_employee_enrollment (read-only): Gets an employee's time tracking enrollment data. - update_employee_enrollment (write, needs approval): Enables, disables, or reassigns a single employee's enrollment using JSON Merge Patch (RFC 7396) semantics: only the properties present in the body are applied and omitted properties are unchanged. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_hour_entries (read-only): Returns a paginated list of time tracking hour entries. - create_hour_entry (write, needs approval): Creates a new time tracking hour entry. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_hour_entry (write, needs approval): Deletes a time tracking hour entry by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_hour_entry (read-only): Retrieves a single hour entry by its ID. - update_hour_entry (write, needs approval): Partially updates a time tracking hour entry identified by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_time_tracking_kiosks (read-only): Returns a paginated list of time tracking kiosks. - delete_time_tracking_kiosk (write, needs approval): Soft-deletes a time tracking kiosk. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_time_tracking_kiosk (read-only): Retrieves a single time tracking kiosk by its ID. - update_time_tracking_kiosk (write, needs approval): Updates a time tracking kiosk using JSON Merge Patch (RFC 7396) semantics. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_time_tracking_projects (read-only): Returns a paginated list of time tracking projects. - create_time_tracking_project (write, needs approval): Creates a time tracking project. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_time_tracking_project (write, needs approval): Deletes a time tracking project by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_time_tracking_project (read-only): Retrieves a single time tracking project by its ID, including the list of employees assigned to it. - update_time_tracking_project (write, needs approval): Partially updates a time tracking project identified by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_time_tracking_project_tasks (read-only): Returns a paginated list of tasks for the specified time tracking project. - create_time_tracking_project_task (write, needs approval): Creates a new task on the specified time tracking project. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_time_tracking_shift_differentials (read-only): Returns a paginated list of time tracking shift differentials. - create_time_tracking_shift_differential (write, needs approval): Creates a new time tracking shift differential. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_time_tracking_shift_differential (write, needs approval): Deletes a time tracking shift differential by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_time_tracking_shift_differential (read-only): Retrieves a single time tracking shift differential by its ID. - update_time_tracking_shift_differential (write, needs approval): Partially updates a time tracking shift differential identified by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_time_tracking_task (write, needs approval): Deletes a time tracking task by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - get_time_tracking_task (read-only): Retrieves a single time tracking task by its ID. - update_time_tracking_task (write, needs approval): Partially updates a time tracking task identified by its ID. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_time_tracking_time_clocks (read-only): Returns a paginated list of the company's time tracking time clocks. - get_time_tracking_time_clock (read-only): Retrieves a single time tracking time clock by its ID. - update_time_tracking_time_clock (write, needs approval): Updates a time tracking time clock using JSON Merge Patch (RFC 7396) semantics: only the properties present in the body are applied. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - approve_timesheet (write, needs approval): Approves a timesheet (process resource). Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_timesheets (read-only): Returns a paginated list of timesheets. - get_timesheet (read-only): Retrieves a single timesheet by its ID. - get_timesheet_summary (read-only): Returns the daily breakdown of hours for a timesheet, including regular, overtime, and double-time hours per day. - delete_timesheet_clock_entries (write, needs approval): Deletes one or more timesheet clock entries by their IDs. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - create_or_update_timesheet_clock_entries (write, needs approval): Creates or updates timesheet clock entries in bulk. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - create_timesheet_clock_in_entry (write, needs approval): Clocks in an employee at the current server time. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - create_timesheet_clock_out_entry (write, needs approval): Clocks out a currently clocked-in employee at the current server time. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_timesheet_hour_entries (write, needs approval): Deletes one or more timesheet hour entries by their IDs. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - create_or_update_timesheet_hour_entries (write, needs approval): Creates or updates timesheet hour entries in bulk. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - create_time_tracking_project_legacy (write, needs approval): Deprecated. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_timesheet_entries (read-only): Returns timesheet entries for all employees, or a filtered subset, within the specified date range. - check_total_rewards_profile_availability (read-only): Check if the given employee has an active Total Rewards profile and the current user has permission to view it. - get_printable_total_rewards_statement (read-only): Generate and return a PDF printable version of the employee's Total Rewards statement. - get_total_rewards_statement (read-only): Returns the full Total Rewards statement for the given employee, including compensation, benefits, bonuses, equity, reimbursements, and time-off data. - list_training_categories (read-only): Returns all training categories for the company as an object keyed by category ID. - create_training_category (write, needs approval): Creates a new training category. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_training_category (write, needs approval): Delete an existing training category. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - update_training_category (write, needs approval): Updates the name of an existing training category. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_employee_training_records (read-only): Returns all training records for the specified employee as an object keyed by training record ID. - create_employee_training_record (write, needs approval): Creates a new training record for the specified employee. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_employee_training_record (write, needs approval): Delete an existing employee training record. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - update_employee_training_record (write, needs approval): Updates an existing employee training record. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - list_training_types (read-only): Returns all training types for the company as an object keyed by training type ID. - create_training_type (write, needs approval): Creates a new training type. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - delete_training_type (write, needs approval): Delete an existing training type. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. - update_training_type (write, needs approval): Updates an existing training type. Writes to BambooHR, so it goes through the confirmation guardrail when enabled. Example prompts: - "List the employee directory (all employees with their basic fields) — from BambooHR." - "Get one employee's details. Pass a comma-separated `fields` list to customize — from BambooHR." - "List time-off requests, optionally filtered by date range and status — from BambooHR." --- ## Basecamp URL: https://www.proxoria.com/en/connectors/basecamp Connect your AI to Basecamp. Basecamp: projects and people, to-do lists and to-dos, messages, comments and Campfire chats, schedules, card tables, documents, automatic check-ins, gauges, timesheets, templates, reports and search — read, and write with confirmation. Authentication: connect Basecamp with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_projects (read-only): List the account's active projects. - get_project (read-only): Get a single project by id (its tools/docks: message board, to-dos, etc.). - list_people (read-only): List all people visible to the account. - get_person (read-only): Get a single person by id. - list_todolists (read-only): List the to-do lists of a project. Resolves the project's to-do set automatically; errors if to-dos are not enabled on that project. Use the returned list ids with list_todos and create_todo. - list_todos (read-only): List the to-dos of a to-do list. Pending by default; pass completed=true for the finished ones. - create_todo (write, needs approval): Create a to-do in a to-do list, optionally with a due date and assignees. Writes to Basecamp, so it goes through the confirmation guardrail when enabled. - complete_todo (write, needs approval): Mark a to-do as completed. Writes to Basecamp, so it goes through the confirmation guardrail when enabled. - get_my_profile (read-only): Get the profile of the authenticated user. - get_account (read-only): Get the Basecamp account itself: name, plan limits, logo. - list_project_people (read-only): List the active people on one project (list_people covers the whole account). - get_out_of_office (read-only): Get a person's out-of-office status. When it is off, enabled is false and no dates are returned. - set_out_of_office (write, needs approval): Turn on (or replace) a person's out-of-office period. People can set their own; setting someone else's needs an admin on a Pro Pack account. - disable_out_of_office (write, needs approval): Turn off a person's out-of-office status. - get_todoset (read-only): Get a project's to-do set — the container of its to-do lists, with completion ratio and counts. Its id is in the project's dock (name 'todoset'). - get_hill_chart (read-only): Get the hill chart of a to-do set: which to-do lists are tracked and where each sits on the hill (figuring it out vs. making it happen). - get_todolist (read-only): Get one to-do list by id: description, completion ratio, and the URLs of its to-dos and groups. Also answers for a to-do list group. - create_todolist (write, needs approval): Create a to-do list in a project's to-do set. - update_todolist (write, needs approval): Rename a to-do list (or group) or change its description. Only the fields you pass change; the rest is kept. - reposition_todolist (write, needs approval): Move a to-do list within its to-do set. Position 1 puts it above every other incomplete list; completed lists are not counted. - list_todolist_groups (read-only): List the groups (sections) inside a to-do list. - create_todolist_group (write, needs approval): Add a group (a named section) to a to-do list. - reposition_todolist_group (write, needs approval): Move a group within its to-do list. - get_todo (read-only): Get one to-do: assignees, dates, completion, its list, and its first 100 subtasks. - create_loose_todo (write, needs approval): Create a to-do directly in a project's to-do set, outside any to-do list. To add one to a list, use create_todo. - update_todo (write, needs approval): Change a to-do: text, notes, assignees, completion subscribers, dates. Only the fields you pass change; pass dueOn or startsOn as null to clear a date. - reposition_todo (write, needs approval): Move a to-do within its list, or into another list of the same project by passing todolistId. - uncomplete_todo (write, needs approval): Reopen a completed to-do. - list_my_assignments (read-only): What is on my plate: my active assignments (to-dos, cards, steps), split into priorities (Up Next) and non_priorities. Not paginated. - list_my_completed_assignments (read-only): My completed assignments (archived and trashed items excluded). - list_my_due_assignments (read-only): My assignments by due date: overdue (the default), today, tomorrow, and so on. - prioritize_assignment (write, needs approval): Add one of my assignments to Up Next. For a card step listed under its card, pass the entry's priority_recording_id. - deprioritize_assignment (write, needs approval): Take an assignment out of Up Next. Removing one that is not there is harmless. - move_priority (write, needs approval): Reorder Up Next: move an already-prioritized assignment to a new position. - list_messages (read-only): List the messages on a project's message board, newest first by default. - get_message (read-only): Get one message with its full content. - create_message (write, needs approval): Post a message on a message board. Without status 'active' it is saved as a draft that notifies no one. Without subscriptions, everyone on the project is notified. - update_message (write, needs approval): Edit a message's subject, body or type, or publish a draft with status 'active'. - list_message_types (read-only): List a project's message types (Announcement, FYI…) for create_message. - get_message_type (read-only): Get one message type of a project. - list_comments (read-only): List the comments on anything in Basecamp — a to-do, message, document, card, upload, schedule entry… — by its id. - get_comment (read-only): Get one comment. - create_comment (write, needs approval): Comment on anything in Basecamp (to-do, message, document, card…). Everyone subscribed to that item is notified. - update_comment (write, needs approval): Replace the text of a comment. - list_campfires (read-only): List every Campfire (project chat) the user can see across the account. - get_campfire (read-only): Get one Campfire. - list_campfire_lines (read-only): Read the lines of a Campfire chat. - get_campfire_line (read-only): Get one line of a Campfire chat. - post_campfire_line (write, needs approval): Say something in a Campfire chat, as the connected user. Plain text unless rich is true. - update_campfire_line (write, needs approval): Edit a Campfire line. Only its author can, only text lines can be edited, and the line becomes rich text. - list_documents (read-only): List the documents in a Docs & Files folder (vault). - get_document (read-only): Get one document with its full content. - list_uploads (read-only): List the uploaded files in a Docs & Files folder (vault), with download URLs. - get_google_document (read-only): Get a linked Google Workspace document (its title, URL and type). - get_schedule (read-only): Get a project's schedule (whether it shows due dates of to-dos and cards). - list_schedule_entries (read-only): List the events on a project's schedule. - get_schedule_entry (read-only): Get one schedule event. For a recurring event, use get_schedule_entry_occurrence to target a given day. - get_schedule_entry_occurrence (read-only): Get the occurrence of a recurring schedule event on a given day. - create_schedule_entry (write, needs approval): Add an event to a project's schedule. For an all-day event pass dates (YYYY-MM-DD) and allDay; otherwise ISO 8601 timestamps. - update_schedule_entry (write, needs approval): Change a schedule event. Only the fields you pass change. A recurring event cannot be changed through the API. - get_upcoming_schedule (read-only): Everything coming up across all projects between two dates: events, occurrences of recurring events, and to-dos, cards and steps that fall due. - get_card_table (read-only): Get a card table (kanban board) with its columns and their card counts. Its id is in the project's dock (name 'kanban_board'). - move_card_column (write, needs approval): Reorder the columns of a card table. Position counts regular columns only (Triage, Not Now and Done are ignored). - create_card_column (write, needs approval): Add a column to a card table. - get_card_column (read-only): Get one column of a card table. - update_card_column (write, needs approval): Rename a card table column or change its description. - set_card_column_color (write, needs approval): Change the color of a card table column. - enable_card_column_on_hold (write, needs approval): Add an On hold section to a card table column. - disable_card_column_on_hold (write, needs approval): Remove the On hold section of a card table column. - watch_card_column (write, needs approval): Start watching a card table column: get notified when cards land in it. - unwatch_card_column (write, needs approval): Stop watching a card table column. - create_card_table_wormhole (write, needs approval): Link a card table to a column of another card table (possibly in another project). The returned id works as a column for move_card: that is how a card changes project. - list_cards (read-only): List the cards in a card table column. - get_card (read-only): Get one card with its content, assignees, due date and steps. - create_card (write, needs approval): Create a card in a card table column. - update_card (write, needs approval): Change a card's title, body, due date or assignees. Only the fields you pass change; dueOn null clears the date, assigneeIds [] removes everyone. - move_card (write, needs approval): Move a card to another column. Passing a wormhole id as the column sends the card to the other card table; that move completes asynchronously and the card gets a new id. - create_card_step (write, needs approval): Add a step (checklist item) to a card. - update_card_step (write, needs approval): Change a card step's title, due date or assignees; the rest is kept. - set_card_step_completion (write, needs approval): Tick or untick a card step. - reposition_card_step (write, needs approval): Move a step within its card. Position 0 is the first step. - list_client_approvals (read-only): List the approvals requested from clients on a project. - get_client_approval (read-only): Get one client approval with its status and responses. - list_client_correspondences (read-only): List the messages exchanged with clients on a project. - get_client_correspondence (read-only): Get one client correspondence. - list_client_replies (read-only): List the client replies to a correspondence or an approval. - get_client_reply (read-only): Get one client reply. - list_gauges (read-only): Project health at a glance: the gauge of every project, riskiest (red) first. - list_gauge_needles (read-only): The history of a project's gauge: every progress update, newest first. - get_gauge_needle (read-only): Get one progress update of a project gauge. - create_gauge_needle (write, needs approval): Record a progress update on a project's gauge: position 0-100 and a color. Notifies nobody unless notify is set. - update_gauge_needle (write, needs approval): Rewrite the description of a gauge update. Position and color cannot change: record a new update instead. - set_project_gauge (write, needs approval): Turn a project's gauge on or off. Project admins only. - get_inbox (read-only): Get a project's Email Forwards inbox. - list_forward_replies (read-only): List the replies to an email forwarded into Basecamp. - get_forward_reply (read-only): Get one reply to a forwarded email. - list_lineup_markers (read-only): List the account-wide markers shown on the Lineup (key dates). - create_lineup_marker (write, needs approval): Put a marker (a named date) on the Lineup, visible to the whole account. - update_lineup_marker (write, needs approval): Rename or move a Lineup marker. - get_my_note (read-only): Read my personal notebook in Basecamp (empty until something is written). - update_my_note (write, needs approval): Replace the content of my personal notebook. Basecamp keeps the previous versions. - get_my_notifications (read-only): My Hey! menu: unread notifications (up to 100), read ones (page by page, 50 each) and bubble-ups. Each item's readable_sgid feeds mark_notifications_read. - list_my_bubble_ups (read-only): My bubble-ups: items I asked to see again, current ones then scheduled ones. - mark_notifications_read (write, needs approval): Mark notifications as read, by the readable_sgid of each (at most 500). - list_my_question_reminders (read-only): The check-in questions still waiting for my answer. - get_questionnaire (read-only): Get a project's Automatic Check-ins tool. - list_questions (read-only): List the check-in questions of a project. - get_question (read-only): Get one check-in question with its schedule. - create_question (write, needs approval): Create a recurring check-in question ("What did you work on today?") that Basecamp asks the project's people on a schedule. - update_question (write, needs approval): Change the wording or the schedule of a check-in question. - pause_question (write, needs approval): Pause a check-in question: Basecamp stops asking it. - resume_question (write, needs approval): Resume a paused check-in question. - update_question_notification_settings (write, needs approval): My notification settings for a check-in question: alert on each answer, and whether the digest lists who has not answered. - list_answers (read-only): Read the answers to a check-in question, optionally for one day or one person. - get_answer (read-only): Get one answer to a check-in question. - list_question_answerers (read-only): List the people who have answered a check-in question. - list_answers_by_person (read-only): Every answer one person gave to a check-in question. - create_answer (write, needs approval): Answer a check-in question as the connected user. - update_answer (write, needs approval): Rewrite an answer to a check-in question; its day is kept unless given. - list_assignable_people (read-only): Everyone who can be assigned to-dos, to then look up their assignments. - get_assigned_todos (read-only): The pending to-dos assigned to one person, grouped by project or by due date. - get_overdue_todos (read-only): All overdue to-dos across projects, grouped by lateness: under a week, over a week, over a month, over three months. - get_progress_report (read-only): The account-wide activity feed: latest events across every project. - get_project_timeline (read-only): The activity feed of one project. - get_person_progress (read-only): What one person has been doing: their activity across projects. - list_templates (read-only): List the project templates, most recent first. - get_template (read-only): Get one project template. - create_template (write, needs approval): Create an empty project template. - update_template (write, needs approval): Rename a project template or change its description. - create_project_from_template (write, needs approval): Start a new project from a template. It is built in the background: poll get_project_construction until its status is completed. - get_project_construction (read-only): Check a project being built from a template; once completed it carries the new project. Poll at most once a second. - get_timesheet_report (read-only): Time logged across the whole account. Without dates, Basecamp returns the last month. - get_project_timesheet (read-only): Time logged on a project, whether against the project itself or its to-dos, cards and messages (each entry's parent says which). - get_recording_timesheet (read-only): Time logged against one item: a to-do, card, message, document… - get_timesheet_entry (read-only): Get one timesheet entry. - create_timesheet_entry (write, needs approval): Log time against an item (to-do, card, message…). For time on the project itself, use the id of the project's timesheet (the parent of its project-level entries). - update_timesheet_entry (write, needs approval): Correct a timesheet entry; only the fields you pass change. - search (read-only): Full-text search across everything the user can see in Basecamp, with filters by type, project, author and period. Example prompts: - "List the account's active projects — from Basecamp." - "Get a single project by id (its tools/docks: message board, to-dos, etc.) — from Basecamp." - "List all people visible to the account — from Basecamp." --- ## Dolibarr URL: https://www.proxoria.com/en/connectors/dolibarr Connect your AI to Dolibarr. Your own Dolibarr ERP/CRM: browse third parties and contacts, sales and purchase documents (invoices with their payments, proposals, orders) in detail, products and stock, projects, tasks, tickets and the agenda, plus warehouses, shipments, contracts, interventions, expense reports, bank accounts, members and users; create third parties, contacts and agenda events, and update a third party. Connects to your instance's REST API with a Dolibarr API key. Authentication: connect Dolibarr with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_thirdparties (read-only): List third parties (companies: customers, prospects, suppliers), optionally filtered by name or by relationship type. - get_thirdparty (read-only): Get one third party by id: address, contact details, VAT number, customer/supplier codes. - create_thirdparty (write, needs approval): Create a third party (company) in the CRM. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - list_contacts (read-only): List contacts (people), optionally filtered by name or by third party. - create_contact (write, needs approval): Create a contact (person), optionally attached to a third party. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - list_invoices (read-only): List customer invoices, optionally restricted to one third party or to a payment status (unpaid is the usual one for chasing payments). - get_invoice (read-only): Get one invoice by id, with its lines, totals and payment situation. - list_proposals (read-only): List commercial proposals (quotes), optionally restricted to one third party. - list_orders (read-only): List sales orders, optionally restricted to one third party. - list_supplier_invoices (read-only): List supplier invoices (what the business owes), optionally restricted to one supplier or to a payment status. - list_supplier_orders (read-only): List purchase orders sent to suppliers, optionally restricted to one supplier. - list_products (read-only): List products and services from the catalog, optionally filtered by label. - list_projects (read-only): List projects (requires the Projects module on the instance). - list_tasks (read-only): List project tasks, optionally those of one project (requires the Projects module). - list_tickets (read-only): List support tickets (requires the Ticket module on the instance). - list_agenda_events (read-only): List agenda events (logged calls, meetings, emails), optionally those of one third party or owned by given users. - create_agenda_event (write, needs approval): Log an agenda event — a call, meeting, email or on-site intervention — optionally linked to a third party. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - get_proposal (read-only): Get one commercial proposal by id, with its lines, totals and validity — the detail list_proposals only summarises. - get_order (read-only): Get one sales order by id, with its lines, totals and delivery date. - get_product (read-only): Get one product or service by id: price, VAT rate, description, barcode and — for a stocked product — the quantity on hand. - list_invoice_payments (read-only): List the payments received against one customer invoice: amount, date and payment type — what tells whether a partly-paid invoice still needs chasing. - update_thirdparty (write, needs approval): Update a third party: contact details, address or VAT number. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - list_records (read-only): List one of the other Dolibarr modules' records: warehouses, stock movements, shipments, contracts, interventions, expense reports, bank accounts, members, users or categories. Each needs its module enabled on the instance; ask for the entity by name. Example prompts: - "List third parties (companies: customers, prospects, suppliers), optionally filtered by name or by relationship type — from Dolibarr." - "Get one third party by id: address, contact details, VAT number, customer/supplier codes — from Dolibarr." - "List contacts (people), optionally filtered by name or by third party — from Dolibarr." --- ## Dropbox Sign URL: https://www.proxoria.com/en/connectors/dropbox-sign Connect your AI to Dropbox Sign. Dropbox Sign (read-only): browse signature requests, their status, and reusable templates. Authentication: connect Dropbox Sign with your API key; credentials are encrypted at rest. This connector is read-only; it never modifies data. Tools: - list_signature_requests (read-only): List signature requests (paginated), with their status. - get_signature_request (read-only): Get a signature request by id (signers, status, documents). - list_templates (read-only): List reusable templates (paginated). Example prompts: - "List signature requests (paginated), with their status — from Dropbox Sign." - "Get a signature request by id (signers, status, documents) — from Dropbox Sign." - "List reusable templates (paginated) — from Dropbox Sign." --- ## ERPNext URL: https://www.proxoria.com/en/connectors/erpnext Connect your AI to ERPNext. Your own ERPNext: read customers, suppliers, items and stock levels, quotations, sales and purchase orders, invoices, delivery notes and receipts, the accounting trail (journal entries, payments, general ledger) and the standard financial reports — trial balance, balance sheet, P&L, ageing — plus projects, tasks, timesheets, employees, contacts and support issues; create customers, leads, opportunities and tasks, and comment on documents. Connects to your instance with an ERPNext API key and secret. Accounting is read-only, and nothing is ever submitted, cancelled or deleted. Authentication: connect ERPNext with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_customers (read-only): List customers, optionally matching a name fragment. - get_customer (read-only): Read one customer by name (the name is the id, from list_customers). - create_customer (write, needs approval): Create a customer. customerType is the party's legal nature (Company, Individual, Partnership); customerGroup is the commercial segment and is a different field. - list_sales_orders (read-only): List sales orders, newest first. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled. - get_sales_order (read-only): Read one sales order with its lines. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled. - list_sales_invoices (read-only): List customer invoices, newest first. Use unpaidOnly for receivables. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled. - get_sales_invoice (read-only): Read one customer invoice with its lines. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled. - list_suppliers (read-only): List suppliers, optionally matching a name fragment. - list_purchase_orders (read-only): List purchase orders, newest first. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled. - get_purchase_invoice (read-only): Read one supplier invoice with its lines. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled. - list_purchase_invoices (read-only): List supplier invoices, newest first. Use unpaidOnly for payables. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled. - list_items (read-only): List items (products and services), optionally matching a name fragment. - get_item (read-only): Read one item by code (the item code is the id). - get_item_stock (read-only): Stock levels for one item, per warehouse. actual_qty is what is physically there; projected_qty also counts what is ordered in and reserved out. An item with no rows has never moved in any warehouse. - list_leads (read-only): List leads, newest first. - create_lead (write, needs approval): Create a lead. - list_opportunities (read-only): List opportunities, newest first. - create_opportunity (write, needs approval): Create an opportunity against an existing customer or lead. opportunityFrom says which of the two party is: pass a customer id with Customer, a lead id with Lead. - list_journal_entries (read-only): List journal entries, newest first. Read-only: this connector never posts, amends or cancels an accounting document. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled. - list_payments (read-only): List payment entries (money in and out), newest first. Read-only. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled. - list_gl_entries (read-only): List general ledger entries — the posted accounting trail, newest first. Cancelled entries are excluded unless includeCancelled is set. Read-only. - add_comment (write, needs approval): Add a comment to a document. Comments are notes on the timeline: they never change the document itself, which is why this is offered on documents that are otherwise read-only. - list_documents (read-only): List one of the other business documents this ERPNext holds: quotations, delivery notes, purchase receipts, stock entries, material requests, projects, tasks, timesheets, employees, contacts, addresses, support issues, work orders or assets. Ask for the entity by name; each has its own useful columns. - get_document (read-only): Read one of those documents in full, including its line items when it has any — a quotation's products, a delivery note's shipped lines, a timesheet's logged hours. - list_reference_data (read-only): List one of the instance's reference tables, to get the values the other tools filter on: warehouses, chart of accounts, cost centres, item groups, customer and supplier groups, territories, units of measure, price lists or companies. - run_financial_report (read-only): Run one of ERPNext's standard accounting reports over a period — trial balance, balance sheet, profit and loss, general ledger, or the receivable/payable ageing. This is the tool for 'where do we stand', rather than reading entries one by one. - create_task (write, needs approval): Create a task, optionally inside a project and with dates. Writes to ERPNext, so it goes through the confirmation guardrail when enabled. Example prompts: - "List customers, optionally matching a name fragment — from ERPNext." - "Read one customer by name (the name is the id, from list_customers) — from ERPNext." - "List sales orders, newest first. docstatus 0 = draft, 1 = submitted (binding, posted to the ledger), 2 = cancelled — from ERPNext." --- ## Eventbrite URL: https://www.proxoria.com/en/connectors/eventbrite Connect your AI to Eventbrite. Manage Eventbrite: list events, view event details, attendees and orders. Authentication: connect Eventbrite with a secure OAuth connection; credentials are encrypted at rest. This connector is read-only; it never modifies data. Tools: - list_events (read-only): List the organization's Eventbrite events. `status` filters by lifecycle state; results are paginated. - get_event (read-only): Get the details of an Eventbrite event. - list_attendees (read-only): List the attendees of an Eventbrite event (paginated). - list_orders (read-only): List the orders (ticket purchases) of an Eventbrite event (paginated). Example prompts: - "List the organization's Eventbrite events. `status` filters by lifecycle state; results are paginated — from Eventbrite." - "Get the details of an Eventbrite event — from Eventbrite." - "List the attendees of an Eventbrite event (paginated) — from Eventbrite." --- ## FreshBooks URL: https://www.proxoria.com/en/connectors/freshbooks Connect your AI to FreshBooks. FreshBooks accounting: browse clients, invoices and expenses, read an invoice, and create invoices. Authentication: connect FreshBooks with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_clients (read-only): List the account's clients. - list_invoices (read-only): List invoices (paginated). - get_invoice (read-only): Get a single invoice by id. - list_expenses (read-only): List expenses (paginated). - create_invoice (write, needs approval): Create an invoice. Pass the FreshBooks invoice object in `invoice` (must include `customerid` and `create_date`; `lines` for line items). Writes to FreshBooks, so it goes through the confirmation guardrail when enabled. Example prompts: - "List the account's clients — from FreshBooks." - "List invoices (paginated) — from FreshBooks." - "Get a single invoice by id — from FreshBooks." --- ## Front URL: https://www.proxoria.com/en/connectors/front Connect your AI to Front. Front shared inbox: read and search conversations, send and reply to messages, comment internally, and manage tags, contacts, drafts, message templates and the knowledge base. Authentication: connect Front with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_conversations (read-only): List conversations (paginated). Use `pageToken` to fetch the next page. - get_conversation (read-only): Get a single conversation by id (status, assignee, tags, recipients). - list_conversation_messages (read-only): List the messages of a conversation (the full thread). - search_conversations (read-only): Search conversations with a Front search query, e.g. 'is:open from:ana@acme.com' or plain words to match the text. Filters are space-separated and ANDed (from/to/cc/bcc OR each other), 15 at most. Supported filters — anything else is rejected as an "unsupported search modifier": is:, from:, to:, cc:, bcc:, recipient:, contact:, participant:, assignee:, author:, mention:, commenter:, inbox:, tag:, link:, custom_field:, before:, after:, during:. `is:` takes open, resolved, archived, snoozed, trashed, assigned, unassigned, unreplied or waiting. inbox:, tag:, link:, assignee:, author:, mention: and commenter: expect Front IDs, not names — resolve them first with list_tags, list_inboxes or list_teammates (tag:TJX fails, tag:tag_16qsy5m works). before/after/during take Unix timestamps. Note the API has NO subject: filter (that one only exists in the Front app); plain words already search the subject and the body. - list_teammates (read-only): List the company's teammates with their ids, names and emails. Needed to search by person: assignee:, author:, mention: and commenter: all take a teammate id (tea_55c8c149), never a name or an email. - list_custom_fields (read-only): List the custom-field definitions of one kind of Front object — their names and ids, which is what a custom_field: search filter needs. - list_accounts (read-only): List the accounts of the company. - get_account (read-only): Fetch an account. - list_account_contacts (read-only): List the contacts associated with an Account. - get_comment (read-only): Fetch a comment. - update_comment (write, needs approval): Update a comment in a conversation. Writes to Front, so it goes through the confirmation guardrail when enabled. - reply_to_comment (write, needs approval): Add a reply to a comment on a conversation. Comment replies visually indicate which comment is being responded to, helping users follow the conversation. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_conversation_comments (read-only): List the comments in a conversation in reverse chronological order (newest first). - comment_conversation (write, needs approval): Add a comment to a conversation. If you want to create a new comment-only conversation, use the Create discussion conversation endpoint. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_contact_notes (read-only): List the notes added to a contact. - add_contact_note (write, needs approval): Create a new note on a contact. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_contacts (read-only): List the contacts of the company (paginated). Use `pageToken` to fetch the next page. - create_contact (write, needs approval): Create a contact at company level. Writes to Front, so it goes through the confirmation guardrail when enabled. - merge_contacts (write, needs approval): Merge several contacts into one. Irreversible: the merged contacts cease to exist as separate records. Writes to Front, so it goes through the confirmation guardrail when enabled. - delete_contact (write, needs approval): Delete a contact. Writes to Front, so it goes through the confirmation guardrail when enabled. - get_contact (read-only): Fetch a contact. - update_contact (write, needs approval): Update a contact. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_contact_conversations (read-only): List the conversations for a contact in reverse chronological order (newest first). For more advanced filtering, see the search endpoint. - list_teammate_contacts (read-only): List the contacts of a teammate. - create_teammate_contact (write, needs approval): Create a contact for a teammate. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_team_contacts (read-only): List the contacts of a team (workspace). - create_team_contact (write, needs approval): Create a contact for a team (workspace). Writes to Front, so it goes through the confirmation guardrail when enabled. - create_discussion (write, needs approval): Create a discussion or task conversation. These only carry comments, never messages — to start an email conversation, send one from a channel with send_message. Writes to Front, so it goes through the confirmation guardrail when enabled. - update_conversation (write, needs approval): Update a conversation. Writes to Front, so it goes through the confirmation guardrail when enabled. - assign_conversation (write, needs approval): Assign a conversation to a teammate, or unassign it by passing no teammate. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_conversation_events (read-only): List what happened on a conversation — assignment, tagging, status changes — newest first. - unlink_conversation (write, needs approval): Detach links from a conversation. The link itself keeps existing, only the attachment to this conversation goes. Writes to Front, so it goes through the confirmation guardrail when enabled. - link_conversation (write, needs approval): Attach one or more links to a conversation — a link points at something outside Front (a deal, a ticket). Writes to Front, so it goes through the confirmation guardrail when enabled. - untag_conversation (write, needs approval): Remove one or more tags from a conversation. Only the tags passed are removed; the others stay. Writes to Front, so it goes through the confirmation guardrail when enabled. - tag_conversation (write, needs approval): Add one or more tags to a conversation. Existing tags are kept. Writes to Front, so it goes through the confirmation guardrail when enabled. - create_draft (write, needs approval): Create a draft as the first message of a NEW conversation, on a channel. Nothing is sent until a human sends it from Front. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_conversation_drafts (read-only): List the drafts in a conversation. - create_draft_reply (write, needs approval): Create a draft reply to the last message of a conversation. Nothing is sent until a human sends it from Front. Writes to Front, so it goes through the confirmation guardrail when enabled. - delete_draft (write, needs approval): Delete a draft. Front requires the draft's current `version` so two editors cannot silently overwrite each other. Writes to Front, so it goes through the confirmation guardrail when enabled. - update_draft (write, needs approval): Edit an existing draft. Front requires the draft's current `version` to detect a concurrent edit. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_inboxes (read-only): List the company's inboxes with their ids. Needed to search by inbox: the search API takes an inbox id (inb_55c8c149), never the inbox name. - get_inbox (read-only): Fetch an inbox. - list_inbox_channels (read-only): List the channels in an inbox. - list_inbox_conversations (read-only): List the conversations in an inbox. For more advanced filtering, see the search endpoint. - list_inbox_teammates (read-only): List the teammates with access to an inbox. - list_teammate_private_inboxes (read-only): List the private inboxes of a teammate. - list_team_inboxes (read-only): List the inboxes belonging to a team (workspace). - delete_article (write, needs approval): Delete an article, with all its content and translations. Writes to Front, so it goes through the confirmation guardrail when enabled. - get_article (read-only): Fetch a knowledge base article. - get_article_content (read-only): Read a knowledge-base article with its body. Pass a locale to read a specific translation rather than the default one. - update_article_content (write, needs approval): Update the body of a knowledge-base article. Pass a locale to write a specific translation rather than the default one. Writes to Front, so it goes through the confirmation guardrail when enabled. - download_article_attachment (read-only): Get the download link of an attachment on a knowledge-base article. - create_article (write, needs approval): Create an article in a knowledge base. Pass a locale to create it in that translation rather than the default one. Writes to Front, so it goes through the confirmation guardrail when enabled. - delete_category (write, needs approval): Delete a knowledge-base category. Warning: every article it contains is deleted with it. Writes to Front, so it goes through the confirmation guardrail when enabled. - get_category (read-only): Fetch a knowledge base category. - list_category_articles (read-only): List articles in a category. - get_category_content (read-only): Read a knowledge-base category with its content. Pass a locale for a specific translation. - update_category_content (write, needs approval): Update a knowledge-base category. Pass a locale to write a specific translation. Writes to Front, so it goes through the confirmation guardrail when enabled. - create_category (write, needs approval): Create a category in a knowledge base. Pass a locale to create it in that translation rather than the default one. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_knowledge_bases (read-only): List the knowledge bases of the company. - create_knowledge_base (write, needs approval): Create a knowledge base. Writes to Front, so it goes through the confirmation guardrail when enabled. - get_knowledge_base (read-only): Fetch a knowledge base. - list_knowledge_base_articles (read-only): List articles in a knowledge base. - list_knowledge_base_categories (read-only): List categories in a knowledge base. - get_knowledge_base_content (read-only): Read a knowledge base with its content. Pass a locale for a specific translation. - update_knowledge_base_content (write, needs approval): Update a knowledge base. Pass a locale to write a specific translation. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_links (read-only): List the company links, filterable by type through `q` (web, app_, jira, asana, monday, trello, github). - get_link (read-only): Fetch a link. For more information on links, see the Links topic. - list_template_folders (read-only): List the message template folders. - create_template_folder (write, needs approval): Create a message template folder in the OLDEST active workspace the token can reach. To choose the workspace, use create_team_template_folder instead. Writes to Front, so it goes through the confirmation guardrail when enabled. - delete_template_folder (write, needs approval): Delete a message template folder and child folders/templates. Writes to Front, so it goes through the confirmation guardrail when enabled. - get_template_folder (read-only): Fetch a message template folder. - update_template_folder (write, needs approval): Update message template folder. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_child_template_folders (read-only): List the folders nested inside a message template folder. - create_child_template_folder (write, needs approval): Create a new message template folder as a child of the given folder. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_teammate_template_folders (read-only): List the message template folders belonging to the requested teammate. - create_teammate_template_folder (write, needs approval): Create a new message template folder belonging to the requested teammate. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_team_template_folders (read-only): List the message template folders belonging to the requested team (workspace). - create_team_template_folder (write, needs approval): Create a new message template folder belonging to the requested team (workspace). Writes to Front, so it goes through the confirmation guardrail when enabled. - list_folder_templates (read-only): List the templates inside a message template folder. - create_folder_template (write, needs approval): Create a new message template as a child of the given folder. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_message_templates (read-only): List the message templates. - create_message_template (write, needs approval): Create a message template in the OLDEST active workspace the token can reach. To choose the workspace, use create_team_message_template instead. Writes to Front, so it goes through the confirmation guardrail when enabled. - delete_message_template (write, needs approval): Delete a message template. Writes to Front, so it goes through the confirmation guardrail when enabled. - get_message_template (read-only): Fetch a message template. - update_message_template (write, needs approval): Update a message template. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_teammate_message_templates (read-only): List the message templates belonging to the requested teammate. - create_teammate_message_template (write, needs approval): Create a new message template for the given teammate. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_team_message_templates (read-only): List the message templates belonging to the requested team (workspace). - create_team_message_template (write, needs approval): Create a new message template for the given team (workspace). Writes to Front, so it goes through the confirmation guardrail when enabled. - send_message (write, needs approval): Send a new email from a channel, which CREATES a new conversation (it supports both messages and comments). To answer inside an existing conversation use reply_to_conversation instead. Writes to Front, so it goes through the confirmation guardrail when enabled. - reply_to_conversation (write, needs approval): Reply inside an existing conversation: the message is sent and appended to the thread. Writes to Front, so it goes through the confirmation guardrail when enabled. - import_message (write, needs approval): Import a message into an inbox WITHOUT sending anything: nothing reaches the recipient. Meant for backfilling history or for sources that do not go through a channel (web forms). Writes to Front, so it goes through the confirmation guardrail when enabled. - get_message (read-only): Fetch one message. Front can also return the raw email (EML) instead of JSON, via the Accept header — this tool asks for JSON. - get_message_seen_status (read-only): Read the seen receipts of a message: one entry per recipient when Front has that detail, otherwise a single entry for the first time it was seen. - list_ticket_statuses (read-only): List the ticket statuses available for your company. - list_company_tags (read-only): List the company tags. - create_company_tag (write, needs approval): Create a company tag. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_tags (read-only): List every tag the token can see — company, team and teammate tags alike — with their ids. Search takes a tag id (tag_16qsy5m), never a name. - create_tag (write, needs approval): Create a tag in the oldest workspace. Front marks this route as legacy: prefer create_company_tag, create_team_tag or create_teammate_tag, which say where the tag belongs. Writes to Front, so it goes through the confirmation guardrail when enabled. - delete_tag (write, needs approval): Delete a tag permanently. It disappears from every conversation it was applied to; the tag itself cannot be restored. Writes to Front, so it goes through the confirmation guardrail when enabled. - get_tag (read-only): Fetch a tag. - update_tag (write, needs approval): Update a tag. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_child_tags (read-only): List the children of a specific tag. - create_child_tag (write, needs approval): Create a child tag. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_tagged_conversations (read-only): List the conversations tagged with a tag. For more advanced filtering, see the search endpoint. - list_teammate_tags (read-only): List the tags for a teammate. - create_teammate_tag (write, needs approval): Create a tag for a teammate. Writes to Front, so it goes through the confirmation guardrail when enabled. - list_team_tags (read-only): List the tags for a team (workspace). - create_team_tag (write, needs approval): Create a tag for a team (workspace). Writes to Front, so it goes through the confirmation guardrail when enabled. - list_teammate_groups (read-only): List the teammate groups in the company. - get_teammate_group (read-only): Fetch a teammate group. - list_teammate_group_inboxes (read-only): List the non-public inboxes explicitly linked to a teammate group (public ones come with its teams). - list_teammate_group_teammates (read-only): List teammate group teammates. - list_teammate_group_teams (read-only): List the teams attached to a teammate group. Attaching a team grants the group access to that team's public inboxes. - get_teammate (read-only): Fetch a teammate. - list_assigned_conversations (read-only): List the conversations assigned to a teammate in reverse chronological order (most recently updated first). For more advanced filtering, see the search endpoint. - list_teammate_inboxes (read-only): Returns list of inboxes the teammate has access to. - list_teams (read-only): List the teams (workspaces) in the company. - get_team (read-only): Fetch a team (workspace). - log_call (write, needs approval): Log a call that happened elsewhere into Front, on a voice channel. This does not place a call — it records one. Writes to Front, so it goes through the confirmation guardrail when enabled. - update_call_status (write, needs approval): Update the status of a call previously logged in Front (ringing, answered, ended…). Writes to Front, so it goes through the confirmation guardrail when enabled. - add_call_summary (write, needs approval): Attach a summary to a call logged in Front. Writes to Front, so it goes through the confirmation guardrail when enabled. - add_call_transcript (write, needs approval): Attach a transcript to a call logged in Front, as a list of lines. Writes to Front, so it goes through the confirmation guardrail when enabled. Example prompts: - "List conversations (paginated). Use `pageToken` to fetch the next page — from Front." - "Get a single conversation by id (status, assignee, tags, recipients) — from Front." - "List the messages of a conversation (the full thread) — from Front." --- ## GoTo Webinar URL: https://www.proxoria.com/en/connectors/goto-webinar Connect your AI to GoTo Webinar. Manage GoTo Webinar: browse webinars, schedule new single-session ones, read registrants and attendees, register or remove participants, and analyse past sessions (performance, polls, questions, surveys). Authentication: connect GoTo Webinar with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_webinars (read-only): List the account's GoTo Webinar sessions within a time range (ISO 8601 dates). - get_webinar (read-only): Get the details of a GoTo Webinar session. - create_webinar (write, needs approval): Schedule a new single-session webinar and return its webinar key. Times are ISO 8601 in UTC (e.g. 2026-09-01T14:00:00Z). Nobody is invited or notified by this call — it only creates the event. Writes to GoTo Webinar, so it goes through the confirmation guardrail when enabled. - list_registrants (read-only): List the people registered for a webinar, with the answers they gave on the registration form (company/organization, job title…) when it collects them. - get_registrant (read-only): Get one registrant's full registration details — including their organization (company), job title and answers to the registration questions, when the form collects them. - register_participant (write, needs approval): Register someone for an upcoming webinar. Returns their registrant key and join URL. Writes to GoTo Webinar, so it goes through the confirmation guardrail when enabled. - remove_registrant (write, needs approval): Cancel someone's registration for a webinar: they lose their join link and drop off the registrant list. This cannot be undone — re-adding them means registering them again, which issues a new join URL. Writes to GoTo Webinar, so it goes through the confirmation guardrail when enabled. - list_attendees (read-only): List who actually attended a past webinar, with their attendance times. Attendees carry no registration answers: to get someone's company, match their `registrantKey` against list_registrants (or call get_registrant). - list_sessions (read-only): List the past sessions (occurrences) of a webinar. A session key is what the performance, polls, questions and surveys tools need. - get_session_performance (read-only): Get a past session's performance: attendance (registrants, attendees, average attention/duration) plus poll and survey summaries. - list_session_polls (read-only): Get the poll questions of a past session with each attendee's answers. - list_session_questions (read-only): Get the questions attendees asked during a past session, with the answers given. - list_session_surveys (read-only): Get the post-webinar survey questions of a past session with their answers. Example prompts: - "List the account's GoTo Webinar sessions within a time range (ISO 8601 dates) — from GoTo Webinar." - "Get the details of a GoTo Webinar session — from GoTo Webinar." - "List the people registered for a webinar, with the answers they gave on the registration form (company/organization, job title…) when it collects them — from GoTo Webinar." --- ## Help Scout URL: https://www.proxoria.com/en/connectors/help-scout Connect your AI to Help Scout. Help Scout: search conversations and read what was actually said in them, look up customers and organizations, leave internal notes and draft replies, manage tags, owners and statuses, and read the full reporting suite — volume, productivity, satisfaction, per agent and for the company. A reply to a customer is written as a draft unless sending is asked for explicitly. Authentication: connect Help Scout with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_mailboxes (read-only): List the mailboxes of the account. - list_conversations (read-only): List conversations, optionally filtered by mailbox and status. - get_conversation (read-only): Get a conversation with its threads (messages). - list_conversation_threads (read-only): List a conversation's threads on their own, paginated — what the customer wrote, what each agent replied, and the internal notes. Use this rather than get_conversation when a long exchange needs reading past its first page. - search_conversations (read-only): Search conversations by their content, subject, customer or tag, using Help Scout's own query syntax — for example `subject:"remboursement"` or `tag:urgent`. Narrower than listing a mailbox and reading through it. - add_note (write, needs approval): Add an internal note to a conversation. Only the team sees it — nothing reaches the customer. This is the safe way for an assistant to leave a summary or a finding on a ticket. - reply_to_conversation (write, needs approval): Write a reply to the customer. It is created as a DRAFT by default, sitting in Help Scout for a human to read and send — because a published reply is emailed to the customer immediately and cannot be recalled. Pass `draft: false` only when the customer has asked for the reply to go out now. For something only the team should see, use add_note. - schedule_reply (write, needs approval): Schedule an existing draft reply to go out at a chosen time. Applies to a draft thread created by reply_to_conversation — take its id from the result. Scheduling again overrides the previous time. - create_conversation (write, needs approval): Open a conversation in a mailbox with a first thread — logging a phone call, recording something a customer said elsewhere. Creating it does not email the customer unless `autoReply` is set. For history brought over from another tool, set `imported` so Help Scout generates no outgoing mail or notification at all. - update_conversation (write, needs approval): Change a conversation's subject, status, owner or mailbox. Assign to a person or a team with `assignTo`, or hand it back to the queue with `unassign`. - set_conversation_tags (write, needs approval): Set the tags of a conversation. THIS REPLACES THE WHOLE SET: every tag missing from the list is removed, and an empty list clears them all. There is no way to add one tag on its own — read the conversation's current tags first with get_conversation, then send them back together with the new one. - list_customers (read-only): List or search customers, by email, name or the organization they belong to. - get_customer (read-only): Get one customer's profile, with their emails, phones and organization. - create_customer (write, needs approval): Create a customer profile. - update_customer (write, needs approval): Update a customer's profile fields. Only what you pass is changed; everything else is left alone. Emails and phones are managed on their own and are not touched here. - get_mailbox (read-only): Get one mailbox, with its sending address and settings. - list_saved_replies (read-only): List a mailbox's saved replies — the wording the team has already agreed on. Worth reading before drafting an answer, so a reply matches how the team actually writes. - list_users (read-only): List the agents of the account. - get_user (read-only): Get one agent by id. - get_current_user (read-only): Get the agent the connection belongs to — who Help Scout thinks is acting. Worth knowing before assigning or signing anything. - list_teams (read-only): List the teams of the account. A team can own a conversation, like an agent can. - list_team_members (read-only): List the agents belonging to a team. - list_tags (read-only): List the tags of the account. Read this before set_conversation_tags, so an existing tag is reused rather than a near-duplicate created. - get_tag (read-only): Get one tag by id. - list_organizations (read-only): List the organizations — the companies customers are grouped under, with their domains, website and notes. - get_organization (read-only): Get one organization by id. - get_rating (read-only): Get one satisfaction rating: the score (great, okay, not_good), the customer's comment, the agent whose reply was rated, and the conversation it belongs to. The id is the rating's own, not the conversation's. - get_company_report (read-only): The company-wide overview for a period: customers helped, conversations closed, replies sent, replies per day and per user, each with the same figures for a previous period and the delta between them. Start here, then drill into whichever number looks wrong. - get_company_customers_helped (read-only): How many distinct customers the team helped, plotted over the period. `viewBy` sets the resolution: day up to 60 days, week up to a year, month beyond 61 days. - get_company_drilldown (read-only): The actual conversations behind a company figure, one page at a time. `range` says which figure — replies, firstReplyResolved, resolved and the rest — and `rangeId` narrows it further where the range takes one. - get_conversations_report (read-only): Conversation volume for a period: how many came in, how many were closed, how many are still open, against the previous period. - get_conversations_new (read-only): New conversations over time — the inbound curve, at the resolution `viewBy` asks for. - get_conversations_busy_times (read-only): When conversations actually arrive, as a grid of weekday against hour. This is the report that answers staffing questions rather than performance ones. - get_conversations_received_messages (read-only): Messages received over time, which is not the same as conversations: one thread can carry many. - get_conversations_volume_by_channel (read-only): Volume split by channel — email, chat, phone — over time. Where the load comes from. - get_conversations_drilldown (read-only): The conversations counted by the conversations report, listed one page at a time. - get_conversations_new_drilldown (read-only): The conversations counted as new, listed one page at a time. - get_conversations_field_drilldown (read-only): The conversations behind one tag, saved reply, workflow or customer. `field` says which of the four, `fieldId` says which one of them. - get_productivity_report (read-only): The team's productivity for a period: first response time, response time, resolution time, replies sent, resolved count, against the previous period. `officeHours` measures elapsed time against the account's business hours instead of the clock. - get_productivity_first_response_time (read-only): First response time over the period — how long a customer waits for the first human reply. - get_productivity_response_time (read-only): Response time over the period, counting every reply and not only the first. - get_productivity_resolution_time (read-only): Resolution time over the period — how long a conversation stays open before it is closed. - get_productivity_replies_sent (read-only): Replies sent over the period, at the resolution `viewBy` asks for. - get_productivity_resolved (read-only): Conversations resolved over the period, at the resolution `viewBy` asks for. - get_happiness_report (read-only): The satisfaction score for a period: how many ratings were great, okay and not good, with the same breakdown for the previous period. - get_happiness_ratings (read-only): The individual satisfaction ratings behind the score, with the customer's comment. `rating` filters to great, ok or not-good. This is where you read what an unhappy customer actually said. - get_user_report (read-only): One agent's figures for a period: conversations handled, replies sent, response and resolution times, happiness, against the previous period. - get_user_customers_helped (read-only): How many distinct customers one agent helped, plotted over the period. - get_user_replies (read-only): One agent's replies sent over the period, at the resolution `viewBy` asks for. - get_user_resolutions (read-only): One agent's resolved conversations over the period, at the resolution `viewBy` asks for. - get_user_happiness (read-only): One agent's satisfaction score for the period, against the previous one. - get_user_ratings (read-only): The individual ratings one agent received, with the customer's comment. Filter with `rating` to read only the bad ones. - get_user_conversation_history (read-only): The conversations one agent handled over the period, listed with their reply count, response time and resolution time. Sort with `sortField` to bring the slowest to the top. - get_user_drilldown (read-only): The conversations behind one agent's figures, listed one page at a time. - get_user_chat_report (read-only): One agent's chat figures for the period: chats handled, response and resolution times. - get_chat_report (read-only): The team's chat figures for the period, against the previous one. - get_email_report (read-only): The team's email figures for the period, against the previous one. - get_phone_report (read-only): The team's phone figures for the period, against the previous one. - get_docs_report (read-only): How the Docs knowledge base performed over the period: views, searches, and the searches that returned nothing — the list of articles your customers wanted and did not find. Example prompts: - "List the mailboxes of the account — from Help Scout." - "List conversations, optionally filtered by mailbox and status — from Help Scout." - "Get a conversation with its threads (messages) — from Help Scout." --- ## HubSpot URL: https://www.proxoria.com/en/connectors/hubspot Connect your AI to HubSpot. HubSpot Forms, contact Lists and CMS (blog posts, site pages, HubDB, URL redirects — not landing pages): browse and search them, read their content, create lists, blog post and site page drafts, edit page content, create URL redirects, and manage multilingual variations of blog posts and site pages. Authentication: connect HubSpot with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_forms (read-only): List marketing forms (paginated). Use `after` to fetch the next page. - get_form (read-only): Get a single marketing form by id, including its field definitions. - list_lists (read-only): Search contact lists by name (leave `query` empty to list them). Returns list ids, names and sizes. - get_list (read-only): Get a single contact list by its ILS list id (name, size, processing type). - list_blog_posts (read-only): List CMS blog posts (paginated). Use `after` to fetch the next page. - get_blog_post (read-only): Get a single CMS blog post by id (title, body, state, slug). - create_blog_post (write, needs approval): Create a CMS blog post. Created as a DRAFT — a human publishes it from HubSpot. Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - update_blog_post (write, needs approval): Update fields of a CMS blog post (title, body, slug, meta description). Does not publish. Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - create_blog_post_translation (write, needs approval): Create a language variation of an existing blog post: clones the source post into the target language and links it into the same translation group (as a draft). Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - set_blog_post_languages (write, needs approval): Relink an existing blog post's multi-language group: set the primary post and map each post id to its language code. Use this to attach posts created separately. Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - list_site_pages (read-only): List CMS site pages (paginated). Landing pages are intentionally not exposed. Use `after` to fetch the next page. - get_site_page (read-only): Get a single CMS site page by id (name, slug, state, html). - list_templates (read-only): List CMS templates in the account (id, label, path, type). Use a template's `path` for create_site_page. Defaults to templates available for new content. - create_site_page (write, needs approval): Create a CMS site page from a template. Created as a DRAFT — a human publishes it from HubSpot. Landing pages are intentionally not exposed. Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - update_site_page (write, needs approval): Update a site page's fields and/or content. Page content lives in the template's modules: pass the FULL layoutSections / widgets / widgetContainers object — HubSpot replaces them wholesale and never merges, so read the page with get_site_page first, edit, then send it back. Does not publish. Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - create_site_page_translation (write, needs approval): Create a language variant of an existing site page: clones the source page into the target language and links it into the same translation group. Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - attach_site_page_translation (write, needs approval): Attach an existing site page to a primary page's multi-language group, as its translation in the given language. Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - detach_site_page_translation (write, needs approval): Detach a site page from its multi-language group (it becomes a standalone page). Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - set_site_page_languages (write, needs approval): Relink an existing site page's multi-language group: set the primary page and map each page id to its language code. Use this to attach pages created separately. Writes to HubSpot, so it goes through the confirmation guardrail when enabled. - list_hubdb_tables (read-only): List HubDB tables (id, name, column definitions, row count). - get_hubdb_table_rows (read-only): List the rows of a HubDB table by table id or name (paginated). - list_url_redirects (read-only): List CMS URL redirects (paginated). Use `after` to fetch the next page. - create_url_redirect (write, needs approval): Create a CMS URL redirect from a source path to a destination. Writes to HubSpot, so it goes through the confirmation guardrail when enabled. Example prompts: - "List marketing forms (paginated). Use `after` to fetch the next page — from HubSpot." - "Get a single marketing form by id, including its field definitions — from HubSpot." - "Search contact lists by name (leave `query` empty to list them). Returns list ids, names and sizes — from HubSpot." --- ## Intercom URL: https://www.proxoria.com/en/connectors/intercom Connect your AI to Intercom. Intercom help center (articles, collections, translations), conversations (read, reply, assign, close) and News announcements: browse and search, create and update content, and act on conversations. Authentication: connect Intercom with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_articles (read-only): List the help-center articles of the workspace (paginated). - get_article (read-only): Get an article with its full content, including existing translations (translated_content). - search_articles (read-only): Search help-center articles by phrase. - list_collections (read-only): List help-center collections (paginated). Their ids are used as parentIds in create_article and set_article_collections. - create_article (write, needs approval): Create a help-center article. Created as a DRAFT by default — a human publishes it from Intercom. Use parentIds to place it in one or more collections/sections. Writes to Intercom, so it goes through the confirmation guardrail when enabled. - update_article (write, needs approval): Update an existing article's default-language fields (title, body, description, state, author). For per-locale translations use update_article_translation; for collection placement use set_article_collections. Publishing only happens if you set state to published. Writes to Intercom, so it goes through the confirmation guardrail when enabled. - set_article_collections (write, needs approval): Place an existing article in one or more collections/sections (replaces its current placement). Writes to Intercom, so it goes through the confirmation guardrail when enabled. - update_article_translation (write, needs approval): Create or update an article's translation for a locale. The translation is saved as a draft unless state is explicitly set to published. - list_conversations (read-only): List conversations, most recent first (paginated). Use startingAfter to page. Good for mining recent support questions (e.g. to draft help-center articles). - get_conversation (read-only): Get a conversation by id, including its message parts. Set plaintext to strip HTML from the message bodies. - reply_conversation (write, needs approval): Reply to a conversation as an admin. messageType 'comment' is sent to the customer; 'note' (default) is an internal note that stays private. Writes to Intercom, so it goes through the confirmation guardrail when enabled. - assign_conversation (write, needs approval): Assign a conversation to an admin or a team. Writes to Intercom, so it goes through the confirmation guardrail when enabled. - close_conversation (write, needs approval): Close a conversation, optionally with a closing note. Writes to Intercom, so it goes through the confirmation guardrail when enabled. - list_newsfeeds (read-only): List the workspace's newsfeeds (their ids are used to publish news items). - list_news_items (read-only): List the workspace's news items (in-app announcements). - get_news_item (read-only): Get a single news item by id (title, body, state, sender). - create_news_item (write, needs approval): Create an in-app News announcement. Created as a DRAFT by default — a human publishes it from Intercom. Writes to Intercom, so it goes through the confirmation guardrail when enabled. - update_news_item (write, needs approval): Update a news item's fields (title, body, sender, labels, state). Making it live only happens if you set state to live. Writes to Intercom, so it goes through the confirmation guardrail when enabled. Example prompts: - "List the help-center articles of the workspace (paginated) — from Intercom." - "Get an article with its full content, including existing translations (translated_content) — from Intercom." - "Search help-center articles by phrase — from Intercom." --- ## Lever URL: https://www.proxoria.com/en/connectors/lever Connect your AI to Lever. Lever recruiting: job postings, candidates and their applications, parsed résumés, files, interview panels and feedback, offers, referrals and notes. Reads the whole hiring record; writes notes, interview feedback and the feedback templates behind it. Authentication: connect Lever with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_postings (read-only): List job postings. Filter by state, team, department, location, work type or tag, or group the results by one of those. Ask for `expand: ['owner', 'hiringManager']` to get the people behind each posting rather than their ids — Lever has no filter by person, so that is how you find someone's postings. - get_posting (read-only): Get one job posting by id, with its content, team, location and owners. - list_posting_users (read-only): List the users who have access to a posting — its owner, hiring manager and followers. - list_opportunities (read-only): List candidates (opportunities). Filter by posting, stage, tag, email, origin or date. `origin: 'referred'` is how you find candidates who came through employee referral. `expand: ['applications', 'stage', 'owner', 'contact']` saves a call per candidate. - get_opportunity (read-only): Get one candidate (opportunity) by id. - list_applications (read-only): List a candidate's applications — one per posting they applied to, with the answers they gave to that posting's questions. - get_application (read-only): Get one application of a candidate in full. - list_resumes (read-only): List a candidate's résumés. Each carries `parsedData`: past positions (company, title, summary, location, dates) and schools (institution, degree, field, dates), already extracted by Lever — so matching a candidate to a posting needs no file download. - get_resume (read-only): Get one résumé of a candidate, with its parsed positions and schools. - list_files (read-only): List the files attached to a candidate — cover letters, portfolios, take-home tests, anything a recruiter dropped on the profile. This returns the metadata only: name, size, type and upload date. Résumés have their own tool, list_resumes, and it is the better one to reach for because Lever parses those into structured positions and schools. - get_file (read-only): Get the metadata of one attached file: its name, extension, size and when it was uploaded. The contents are not returned — Lever serves those from a separate download route this connector does not expose. - list_offers (read-only): List the offers made to a candidate: status, who created it, when, the salary and the other fields the offer form captured. The status is how you tell a candidate who was offered from one who signed — `signed` means accepted. Lever publishes no account-wide list of offers, so this is per candidate. - list_feedback (read-only): List the interview feedback forms filled in for a candidate, with every field and score. This is where an interview panel's actual judgement lives. - get_feedback (read-only): Get one feedback form filed on a candidate, with every field, the value the interviewer chose and when they completed it. - list_feedback_templates (read-only): List the feedback form templates of the account, with each field and its id. Required before create_feedback: the field ids of the values you submit come from here. - get_feedback_template (read-only): Get one feedback template with its fields, their type, their options and which stage it belongs to. Read it before update_feedback_template: that call replaces the template, so what you do not send back is what you lose. - update_feedback_template (write, needs approval): Rewrite a feedback template. This is account-wide and forward-looking: every interviewer who opens this form from now on sees what you leave here, so it is a change to the hiring process, not to one candidate. Lever replaces the template with what you send, so read get_feedback_template first and send back the fields you want to keep. Exactly one field must have type `score-system` — that is the overall rating Lever reports on. - create_feedback (write, needs approval): File a completed interview feedback form on a candidate. Lever records it under the user named in `performAs` and shows no sign that it was not typed by them — so only use this for someone who has told you what to record, never to invent an assessment. Take `baseTemplateId` and every field id from list_feedback_templates. - update_feedback (write, needs approval): Change the answers on a feedback form already filed on a candidate. Lever records the edit under the user named in `performAs` and shows no sign that it was not typed by them — so only use this for someone who has told you what to change, never to reword another interviewer's assessment. Only the fields you send are touched; take their ids from get_feedback. Setting `completedAt` is what marks a form as done. - list_interviews (read-only): List the interviews scheduled for a candidate, with their panel and interviewers. - get_interview (read-only): Get one scheduled interview: its panel, date, duration, location, the interviewers and the feedback form each of them owes. `feedbackForms` lists the forms already filed for it — an empty one is an interview nobody has written up yet. - list_panels (read-only): List the interview panels of a candidate. A panel is one scheduled block — an on-site day, a phone screen — holding the interviews it is made of, their date and their interviewers. Use it to see the shape of a candidate's process; use list_interviews for the individual slots. - get_panel (read-only): Get one interview panel with every interview it contains, its stage, its timezone and who cancelled it if it was cancelled. - list_referrals (read-only): List the referral forms filled in for a candidate. The referrer appears as a form field (often 'Name of referrer'), not as a Lever user — so counting who refers most means grouping on a typed name, and two spellings of one person will not merge. Find the candidates worth asking about with list_opportunities and `origin: 'referred'`; Lever publishes no account-wide list of referrals. - get_referral (read-only): Get one referral form with every field the referrer filled in, and when they filled it. - list_opportunity_notes (read-only): List the notes left on a candidate — the running commentary recruiters and managers add outside the formal feedback forms. - get_opportunity_note (read-only): Get one note thread on a candidate, with each of its comments, who wrote them and when. - add_note (write, needs approval): Add a note to a candidate. Notes are visible to everyone with access to the candidate. Give `performAs` to record it under a specific user; without it Lever attributes the note to the account the API key belongs to. A note added here can be rewritten later with update_note, which is not true of notes typed into Lever itself. - update_note (write, needs approval): Rewrite a note on a candidate. Two things make this narrower than it looks. Lever only accepts it on notes that were created through the API, so a note somebody typed into Lever cannot be edited here and the call is refused. And it replaces the whole thread: every comment you want to keep must be sent again, so read get_opportunity_note first and send back what should survive. - list_stages (read-only): List the pipeline stages of the account. - get_stage (read-only): Get one pipeline stage by id — its name, so a stage id coming back from a candidate or an interview can be read as the step it actually is. - list_users (read-only): List the Lever users of the account — recruiters, hiring managers, interviewers. Their ids are what performAs and the posting owners refer to. - list_archive_reasons (read-only): List the reasons an application can be archived for — why candidates were turned down, and which of those reasons count as hired. Example prompts: - "List job postings. Filter by state, team, department, location, work type or tag, or group the results by one of those. Ask for `expand: ['owner', 'hiringManager']` to get the people behind each posting rather than their ids — Lever has no filter by person, so that is how you find someone's postings — from Lever." - "Get one job posting by id, with its content, team, location and owners — from Lever." - "List the users who have access to a posting — its owner, hiring manager and followers — from Lever." --- ## LinkedIn URL: https://www.proxoria.com/en/connectors/linkedin Connect your AI to LinkedIn. Publish posts on LinkedIn as the connected member — text, image, several images, video, PDF document, poll, article link or a repost — and edit or delete them. Authentication: connect LinkedIn with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - get_profile (read-only): Get the connected LinkedIn member's profile (name, locale, picture). - create_post (write, needs approval): Publish a post on LinkedIn as the connected member — it goes live immediately (LinkedIn has no draft mode). Attach at most ONE of: an image, 2-20 images, a video, a PDF/PPT/DOC document, an article link, or a poll; or repost an existing post. Media is given as an HTTPS URL (max 10 MB) and uploaded to LinkedIn for you. LinkedIn allows a member 150 requests a day and a post with media costs three, so pace bulk runs. - update_post (write, needs approval): Edit one of the connected member's own published posts. LinkedIn only allows the text and the call-to-action to change — the attached image, video or document cannot be swapped; delete and repost for that. You need the post URN that create_post returned: LinkedIn does not let this connector list a member's posts back. - delete_post (write, needs approval): Delete one of the connected member's own posts. This is permanent and takes the post's comments and reactions with it. Example prompts: - "Get the connected LinkedIn member's profile (name, locale, picture) — from LinkedIn." --- ## LinkedIn Community Management URL: https://www.proxoria.com/en/connectors/linkedin-community-management Connect your AI to LinkedIn Community Management. Manage LinkedIn company pages the connected member administers: publish and edit posts (text, image, video, PDF document, multi-image, poll, article, repost), moderate comments and reactions, and read page, follower, post and video analytics. Authentication: connect LinkedIn Community Management with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_organizations (read-only): List the company pages the connected member administers, with their id and name. Use the id with create_post. If this returns nothing, the member holds no ADMINISTRATOR role on any page — the page id can also be read from the page's admin URL. - get_organization (read-only): Get one company page's details (name, vanity name, description, website, industry, staff count). Look it up by page id, by vanity name (the last part of its LinkedIn URL), or by the email domain its employees use. - list_organization_brands (read-only): List the brand pages (showcase pages) attached to a parent company page. Each brand has its own id and can be posted to and measured like a page. - list_organization_admins (read-only): List the members who hold a role on a company page (administrators, direct sponsored content posters…), with their role and whether it is approved. Returns member URNs — LinkedIn does not expose admin names through this API. - get_follower_count (read-only): Get the current total number of followers of a company page. For the breakdown by seniority, function, industry or country, and for gains over time, use get_follower_statistics instead. - create_post (write, needs approval): Publish a post on a LinkedIn company page — it goes live immediately (LinkedIn has no draft mode). Attach at most ONE of: an image, 2-20 images (carousel of images), a video, a PDF/PPT/DOC document, an article link, or a poll; or repost an existing post. Media is given as an HTTPS URL (max 10 MB) and uploaded to LinkedIn for you. - list_posts (read-only): List the posts published by a company page, newest first, with their text, type and URN. Use the returned URN with get_post, update_post, list_comments or get_share_statistics. - get_post (read-only): Get one post by its URN (or its LinkedIn URL), with its text, media, visibility and lifecycle state. - update_post (write, needs approval): Edit a published post. LinkedIn only allows the text and the call-to-action to change — the attached image, video or document cannot be swapped; delete and repost for that. - delete_post (write, needs approval): Delete a post from the company page. This is permanent and takes the post's comments and reactions with it. - list_comments (read-only): List the comments on a post, or the replies under a comment. Each comment carries its author URN, text and id — pass a comment URN back here to read its replies. - create_comment (write, needs approval): Comment on a post as the company page, or reply to an existing comment. The text is posted as plain text — hashtags and @mentions are not linked from here. - delete_comment (write, needs approval): Delete a comment the company page made, or a comment on the page's own post. Permanent — LinkedIn keeps no copy. - list_reactions (read-only): List who reacted to a post or a comment, and with which reaction. For the totals per reaction type, get_social_metadata is one call instead of paging through these. - create_reaction (write, needs approval): React to a post or comment as the company page. Reaction types: LIKE, PRAISE (Celebrate), EMPATHY (Love), INTEREST (Insightful), APPRECIATION (Support), ENTERTAINMENT (Funny). - delete_reaction (write, needs approval): Remove the company page's reaction from a post or a comment. - get_social_metadata (read-only): Get the engagement summary of a post or comment in one call: the count of each reaction type, the number of comments and replies, and whether the thread is open to comments. - set_comments_enabled (write, needs approval): Open or close a post to comments. Closing it DELETES every existing comment on that post — reactions are kept, and reopening does not bring the comments back. - get_page_statistics (read-only): Get a company page's view and click statistics, broken down by page section (overview, jobs, people…). Omit the dates for lifetime totals; give them for a period. - get_follower_statistics (read-only): Get a company page's follower statistics. Without dates: the lifetime breakdown by seniority, function, industry, company size, country and organic vs paid. With dates: followers gained per day or month over the period. - get_share_statistics (read-only): Get organic post performance for a company page: impressions, unique impressions, clicks, likes, comments, shares and engagement rate. Aggregated for the page, per day or month over a period, or per post when you pass postUrns. Covers the last 12 months only, and excludes sponsored activity. - get_video_analytics (read-only): Get the viewing figures of a video post on a company page: video views, unique viewers or time watched. LinkedIn keeps these for six months after the post. - search_followers (read-only): Find members among a company page's followers — by keyword on their name, or by their LinkedIn profile URL. Returns member URNs, which is what an @mention in a post needs. - get_member_follower_count (read-only): Get the follower count of the connected member's own LinkedIn profile (not a company page). Omit the dates for the current total, or give them for the gain over a period. - get_member_post_analytics (read-only): Get analytics on the connected member's OWN posts (their personal profile, not a company page): impressions, members reached, reactions, comments, reposts, saves, link clicks and followers gained. Give a post URN for one post, or omit it for the member's posts as a whole. Example prompts: - "List the company pages the connected member administers, with their id and name. Use the id with create_post. If this returns nothing, the member holds no ADMINISTRATOR role on any page — the page id can also be read from the page's admin URL — from LinkedIn Community Management." - "Get one company page's details (name, vanity name, description, website, industry, staff count). Look it up by page id, by vanity name (the last part of its LinkedIn URL), or by the email domain its employees use — from LinkedIn Community Management." - "List the brand pages (showcase pages) attached to a parent company page. Each brand has its own id and can be posted to and measured like a page — from LinkedIn Community Management." --- ## LinkedIn Marketing URL: https://www.proxoria.com/en/connectors/linkedin-marketing Connect your AI to LinkedIn Marketing. Run LinkedIn marketing for the ad accounts the connected member administers: ad accounts, campaign groups, campaigns, creatives, dark posts (Direct Sponsored Content) and audience targeting, plus performance and demographic reporting. Authentication: connect LinkedIn Marketing with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_ad_accounts (read-only): List the LinkedIn ad accounts the connected member can act on, with their id, name, currency and status. The id feeds every other tool here. - get_ad_account (read-only): Get one ad account in detail: name, currency, type, status, the organization it bills to and its notification settings. - list_ad_account_users (read-only): List the members who have a role on an ad account (billing admin, account manager, campaign manager, creative manager, viewer) — useful to check who can approve a change. - list_campaign_groups (read-only): List an ad account's campaign groups with their budget, schedule and status. A campaign group is the container that holds campaigns and can cap their combined spend. - create_campaign_group (write, needs approval): Create a campaign group in an ad account. Created PAUSED unless you pass status ACTIVE, so nothing starts spending by accident. - update_campaign_group (write, needs approval): Change a campaign group's name, budget, schedule or status. Setting status to ACTIVE lets its campaigns start delivering; PAUSED stops them. - list_campaigns (read-only): List an ad account's campaigns with their type, objective, budget, schedule and status. Optionally narrow to one campaign group. - get_campaign (read-only): Get one campaign in full, including its targeting criteria, bid and budget settings. - create_campaign (write, needs approval): Create a campaign with its budget, schedule and audience targeting. Give targeting as lists of URNs discovered with search_targeting_entities — facets intersect (a location AND a seniority) while URNs within one facet are alternatives. Created as a DRAFT that spends nothing until you activate it with update_campaign. LinkedIn tools may not be used to discriminate based on personal characteristics like gender, age, race, or ethnicity. - update_campaign (write, needs approval): Change a campaign's name, budget, bid, schedule or status. This is how a DRAFT campaign goes live: set status to ACTIVE. Setting it to PAUSED stops delivery and spend. - list_creatives (read-only): List an ad account's creatives — the ads themselves — with the post each one promotes and its review status. Optionally narrow to one campaign. - get_creative (read-only): Get one creative: the content it promotes, its campaign, status and review state. - create_creative (write, needs approval): Turn an existing LinkedIn post into an ad by attaching it to a campaign. Give the URN of a post already published by the company page (list it with the LinkedIn Community Management connector). Created PAUSED unless you say otherwise. - update_creative (write, needs approval): Activate, pause or archive a creative — this is how an ad starts or stops running. - delete_creative (write, needs approval): Delete a creative. Only a creative that never ran can be deleted — pause or archive one that has already been served. - create_dark_post (write, needs approval): Create a dark post — Direct Sponsored Content: an ad authored by a company page that NEVER appears on the page's feed and reaches nobody until a campaign pays to show it. This is how you run several message or visual variants of the same offer without flooding the page with near-identical posts. Attach it to a campaign with create_creative afterwards, using the postId this returns. - list_dark_posts (read-only): List the dark posts (Direct Sponsored Content) belonging to an ad account. Worth using before creating another: dark posts are invisible on the company page, so this finder is the only way to see what already exists and avoid leaving orphans behind. - list_targeting_facets (read-only): List the targeting categories LinkedIn offers (locations, industries, job functions, seniorities, skills, company size…) and which search method each supports. Start here, then use search_targeting_entities to find the URNs inside a facet. LinkedIn tools may not be used to discriminate based on personal characteristics like gender, age, race, or ethnicity. - search_targeting_entities (read-only): Find the targeting URNs a campaign needs. Search by keyword within a facet (best for locations, industries, titles, skills, employers), list a whole facet (best for seniorities, job functions, company sizes), find entities similar to ones you have, or resolve URNs back to names. LinkedIn tools may not be used to discriminate based on personal characteristics like gender, age, race, or ethnicity. - get_audience_count (read-only): Estimate how many LinkedIn members a set of targeting criteria reaches, before creating a campaign. LinkedIn requires an audience of at least 300 members to deliver. - get_ad_analytics (read-only): Report on advertising performance: impressions, clicks, spend, conversions, leads and engagement. Break the figures down by campaign, creative, campaign group — or by audience demographics (member company, industry, job function, seniority, job title, company size, country) to see who actually responded. - get_supply_forecast (read-only): Forecast the reach and impressions a budget and audience would buy, before committing to a campaign. Useful to sanity-check a plan against a bid level. Example prompts: - "List the LinkedIn ad accounts the connected member can act on, with their id, name, currency and status. The id feeds every other tool here — from LinkedIn Marketing." - "Get one ad account in detail: name, currency, type, status, the organization it bills to and its notification settings — from LinkedIn Marketing." - "List the members who have a role on an ad account (billing admin, account manager, campaign manager, creative manager, viewer) — useful to check who can approve a change — from LinkedIn Marketing." --- ## Livestorm URL: https://www.proxoria.com/en/connectors/livestorm Connect your AI to Livestorm. Livestorm webinars: browse events and their sessions, list the people registered for a session (registrants and attendees), and register a participant. Authentication: connect Livestorm with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_events (read-only): List the account's webinar events (paginated). - get_event (read-only): Get a single webinar event by id (title, settings, registration page). - list_sessions (read-only): List the sessions (scheduled occurrences) of a webinar event (paginated). - list_session_people (read-only): List the people of a session — registrants and attendees, with their attendance status (paginated). - register_participant (write, needs approval): Register a participant for a session (creates a registrant). Writes to Livestorm, so it goes through the confirmation guardrail when enabled. Example prompts: - "List the account's webinar events (paginated) — from Livestorm." - "Get a single webinar event by id (title, settings, registration page) — from Livestorm." - "List the sessions (scheduled occurrences) of a webinar event (paginated) — from Livestorm." --- ## Lucca API V3 URL: https://www.proxoria.com/en/connectors/lucca-v3 Connect your AI to Lucca API V3. Lucca (legacy v3 API): directory, absences (Timmi), expenses (Cleemy), timesheets (Timmi) and office locations; plus leave approval and user create/update. Authentication: connect Lucca API V3 with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_users (read-only): List users from the directory. Optionally filter by email and include former employees. - get_user (read-only): Get one user's details. Pass a comma-separated `fields` list to customize. - list_departments (read-only): List the organization's departments. - list_leave_requests (read-only): List leave requests (Timmi Absences), pending or already decided. - list_leaves (read-only): List actually-booked leaves over a date range (Timmi Absences). - decide_leave_request (write, needs approval): Approve or deny a pending leave request. A comment is required when denying. - list_expense_claims (read-only): List Cleemy expense claims, optionally filtered by owner and status. - list_timesheets (read-only): List timesheets (Timmi), optionally filtered by owner and date range. - list_time_entries (read-only): List time entries (Timmi), optionally filtered by owner and date range. - list_user_locations (read-only): List where employees are working (Timmi Office) over a date range. - list_work_locations (read-only): List the available work locations / sites (Timmi Office). - create_user (write, needs approval): Create a user in the directory. Pass the full Lucca user object in `user` (Lucca requires firstName, lastName, login, mail, dtContractStart, legalEntityId, employeeNumber, departmentId, managerId, rolePrincipalId, habilitedRoles, birthDate, calendarId, cultureID). - update_user (write, needs approval): Update a user (full replace — PUT). Pass the complete Lucca user object in `user`; omitted required fields will be cleared, so include all of them. - list_expenseclaimitems (read-only): Retrieve a list of `expenseClaimItems`. - get_expenseclaimitem (read-only): Retrieve an expense claim item by its identifier. - create_expenseclaim (write, needs approval): When a user wants to declare his expenses, he creates an `ExpenseClaim`. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_expensetempitems (read-only): Retrieve a list of `ExpenseTempItems`. - create_temporary_expense (write, needs approval): Create a new temporary-expense-item. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_expensetempitem (read-only): Retrieve an temporary expense item by its identifier. - update_expensetempitem (write, needs approval): Update a temporary expense item by its identifier. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - create_expense_accounting_export (write, needs approval): Create a new export and retrieve the export content directly in the response body for a given business-establishment. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_work_contracts (read-only): Retrieve a paginated collection of work-contracts. - get_work_contract (read-only): Retrieve a single work-contract by its ID. - list_booked_documents (read-only): List booked-documents (i.e. - create_commitment_request (write, needs approval): Trigger the approval workflow for a purchase commitment. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_cost_centers (read-only): List cost-centers. - create_cost_center (write, needs approval): Create a new cost-center. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_cost_center (read-only): Retrieve a cost-center by its ID. - update_cost_center (write, needs approval): Update a cost-center. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - delete_document (write, needs approval): Delete a document. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_document (read-only): Retrieve a document by its ID. - update_document (write, needs approval): Update a document's information (e.g. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_invoice (read-only): Retrieve an invoice by its ID. - update_invoice (write, needs approval): Update an invoice. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_invoice_confirmations (read-only): List confirmations for an invoice. - confirm_invoice (write, needs approval): Confirm (approve) an invoice. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_natures (read-only): List natures. - create_nature (write, needs approval): Create a new nature. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_nature (read-only): Retrieve a nature by its ID. - update_nature (write, needs approval): Update a nature. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_purchases (read-only): List purchases. - create_purchase (write, needs approval): Create a Purchase. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_purchase (read-only): Retrieve a purchase by its ID. - close_purchase (write, needs approval): Close a purchase. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - create_commitment (write, needs approval): Create a commitment. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - cancel_commitment_request (write, needs approval): Cancel a commitment-request. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_suppliers (read-only): List suppliers. - create_supplier (write, needs approval): Create a new supplier. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_supplier (read-only): Retrieve a supplier by its ID. - update_supplier (write, needs approval): Update a supplier. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - import_cost_centers (write, needs approval): Import a list of cost-centers. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_inbox_documents (read-only): List documents waiting to be processed in the inbox. - update_inbox_document (write, needs approval): Update an inbox document and optionally attach or create a purchase. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - book_inbox_document (write, needs approval): Book (validate) a document from the inbox. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_vat_rates (read-only): List V.A.T. - list_department_tree (read-only): List all departments as a tree. - get_department (read-only): Retrieve a single Department by its unique identifier. - list_structure_departments (read-only): List departments. - get_structure_department (read-only): Retrieve a department by its ID. - list_establishments (read-only): Retrieve a list of establishments. - list_legal_units (read-only): Retrieve a paginated list of legal-units. - list_fixed_items (read-only): Retrieve fixed remuneration items matching a set of filters. - list_fixed_nature_categories (read-only): List fixed-nature categories. - list_fixed_natures (read-only): Retrieve all existing fixed natures. - get_fixed_nature (read-only): Retrieve a single fixed nature from its identifier. - list_variable_items (read-only): Retrieve variable remuneration items matching a set of filters. - create_variable_item (write, needs approval): Create a new variable item attached to an employee work contract. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - delete_variable_item (write, needs approval): Delete a variable-item. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_variable_item (read-only): Retrieve a single variable item by id. - update_variable_item (write, needs approval): Update a variable-item. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_variable_nature_categories (read-only): List variable-nature-categories. - list_variable_natures (read-only): Retrieve all existing variable natures. - get_variable_nature (read-only): Retrieve a single variable nature from its identifier. - list_realized_trainings (read-only): Retrieve a paginated list of realized-trainings for all employees who were present at a closed session. - list_training_categories (read-only): Retrieve a paginated list of training-categories. - list_training_providers (read-only): Retrieve a paginated list of training-providers. - list_trainings (read-only): Retrieve a paginated list of trainings. - create_training (write, needs approval): Creates a training with the state 'ready', which by default is not displayed in the public catalog. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_training (read-only): Retrieve a training. - list_realized_trainings_v2 (read-only): Retrieve a paginated list of realized trainings. - create_realized_training_v2 (write, needs approval): Create a new realized training. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - delete_realized_training_v2 (write, needs approval): Delete a realized training by its id. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_realized_training_v2 (read-only): Retrieve a single realized-training by its id. - list_training_demands (read-only): Retrieve a paginated list of training-demands that satisfy the query filters. - list_user_dates (read-only): Retrieve a list of UserDates. - list_workcycleexceptions (read-only): Retrieve a paginated list of WorkCycleExceptions that satisfy the given query filters. - get_workcycleexception (read-only): Retrieve a single WorkCycleException by its ID. - list_employee_assignments (read-only): List Employee Assignments. You may not benefit from this feature. - list_shiftoverrides (read-only): Retrieve a paginated list of ShiftOverrides that satisfy the given query filters. - list_working_time_arrangements (read-only): List Working Time Arrangements. You may not benefit from this feature. - get_working_time_arrangement (read-only): Get a working time arrangement by id. Beta feature at Lucca: the account may not have it. - list_collective_schedules_from_a_working_time_arrangement (read-only): List the collective schedules of a working time arrangement. Beta feature at Lucca: the account may not have it. - get_collective_schedule_from_a_working_time_arrangement (read-only): Get one collective schedule of a working time arrangement. Beta feature at Lucca: the account may not have it. - get_leave_request (read-only): Retrieve a single Leave Request by its unique identifier. - get_leave (read-only): Retrieve a single Leave by its unique identifier. - list_sickleavecertificates (read-only): Retrieve a paginated list of SickLeaveCertificate objects, which contain further information about certain types of leaves. - get_sickleavecertificate (read-only): Retrieve a single SickLeaveCertificate. - get_import_user_locations_analysis (read-only): Retrieve the analysis of the Import User Locations API request with the create parameter set to false. - get_import_user_locations_progress (read-only): Retrieve the progress of the Import User Locations API request. - get_import_user_locations_results (read-only): Retrieve the results of the Import User Locations API request with the create parameter set to true. - get_specific_userlocation (read-only): Get specific userLocation by id. - get_specific_worklocation (read-only): Get a specific work location by id. - list_organizations (read-only): List organizations. - list_project_financials (read-only): Retrieve all financial metrics on projects. - list_clients (read-only): List clients from an organization. - create_client (write, needs approval): Create a new client. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - delete_client (write, needs approval): Delete a single client. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_client (read-only): Retrieve a single client identified by its unique identifier. - update_client (write, needs approval): Update fields of a single client identified by its unique id. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_projects (read-only): List projects from an organization. - create_project (write, needs approval): Create a new Project. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - delete_project (write, needs approval): Only projects at "draft" status can be deleted. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_project (read-only): Retrieve a project by its unique identifier. - update_project (write, needs approval): Update an existing Project. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_project_services (read-only): List all services from a project identified by its unique id. - create_project_service (write, needs approval): Create a new Project Service on a project. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - approve_timesheets (write, needs approval): Approve once each timesheet referenced by their ID, and any timesheets before each of them pending approval belonging to the same timesheet owner. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - submit_timesheets (write, needs approval): Submit all timesheets for an owner up to the given period, or to the last submittable timesheet. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - approve_timesheet (write, needs approval): Approve once the timesheet referenced by its ID, and any timesheets before it pending approval belonging to the same timesheet owner. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - cancel_timesheet (write, needs approval): Cancel a timesheet referenced by its ID, and any timesheets after it pending approval belonging to the same timesheet owner. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - deny_timesheet (write, needs approval): Deny a timesheet with a mandatory comment. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - invalidate_timesheet (write, needs approval): Invalidate a timesheet with a mandatory comment. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - update_time_entries_v4 (write, needs approval): ## Before using this service This service is aimed at updating a given user TimeEntries easily, as it automatically detects and applies changes needed to transform whatever TimeEntries might…. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - delete_multiple_time_entries (write, needs approval): Delete one or several time-entries. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - create_time_entries (write, needs approval): Create one or multiple time-entries. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - update_multiple_time_entries (write, needs approval): Update one or several time-entries. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - delete_time_entry (write, needs approval): Delete a time-entry. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - get_time_entry (read-only): Retrieve a single time-entry identified by its unique identifier. - update_time_entry (write, needs approval): Update fields of a single time-entry identified by its unique id. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - list_due_timesheets (read-only): List timesheets that are not yet submitted (status: 0). - create_report (write, needs approval): This endpoint does not adhere to the "v3 API endpoints" principles. Writes to Lucca, so it goes through the confirmation guardrail when enabled. - download_report_as_csv (read-only): This endpoint does not adhere to the "v3 API endpoints" principles. - download_report_as_excel (read-only): This endpoint does not adhere to the "v3 API endpoints" principles. - remind_timesheets (write, needs approval): Remind users of a due timesheet. Writes to Lucca, so it goes through the confirmation guardrail when enabled. Example prompts: - "List users from the directory. Optionally filter by email and include former employees — from Lucca API V3." - "Get one user's details. Pass a comma-separated `fields` list to customize — from Lucca API V3." - "List the organization's departments — from Lucca API V3." --- ## Lucca API V5 URL: https://www.proxoria.com/en/connectors/lucca-v5 Connect your AI to Lucca API V5. Lucca on the v5 API (read-only, apart from departments): people and their personal records, contracts and probation, job positions and qualifications, the organisation chart, and the custom attributes and taxonomies a company defines. Bank and insurance details are never returned. Authentication: connect Lucca API V5 with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_employees (read-only): List employees from the directory. - get_employee (read-only): Get one employee's details by id. - list_departments (read-only): List the organization's departments. - list_establishments (read-only): List the company's business establishments (sites). - list_legal_entities (read-only): List the company's legal entities. - list_personal_records (read-only): List the employees' personal records. Bank details (bankAccount) and national insurance numbers (insuranceNumber) are NEVER returned by this connector: the field is present but replaced by a notice. Everything else is returned — civil title, legal gender, birth date, nationality, personal phone and personal email. - get_personal_record (read-only): Get one employee's personal record. As with the list, bankAccount and insuranceNumber are never returned — do not report them as missing data, they are deliberately withheld. - list_employments (read-only): List employment contracts, with their dates, working time and the job they attach to. - get_employment (read-only): Get one employment contract in full. - list_employment_templates (read-only): List the contract templates the company works from. - get_employment_template (read-only): Get one contract template. - list_probationary_periods (read-only): List probationary periods, with their end dates — what tells you whose trial period is about to close. - get_probationary_period (read-only): Get one probationary period. - list_working_time_arrangements (read-only): List the working-time arrangements — full time, part time, and the local variants a company defines. - get_working_time_arrangement (read-only): Get one working-time arrangement. - list_job_positions (read-only): List the job positions — the posts people are appointed to. - get_job_position (read-only): Get one job position. - list_professions (read-only): List the professions the company recognises. - get_profession (read-only): Get one profession. - list_job_qualifications (read-only): List the job qualifications — the classification levels a contract refers to. - get_job_qualification (read-only): Get one job qualification. - list_occupation_categories (read-only): List the occupation categories — the socio-professional bands a payroll system depends on. - get_occupation_category (read-only): Get one occupation category. - get_legal_entity (read-only): Get one legal entity. - get_establishment (read-only): Get one business establishment. - get_department (read-only): Get one department, with its place in the hierarchy. - list_employee_attributes (read-only): List the values of the custom attributes carried by employees — the company's own fields. - get_employee_attribute (read-only): Get one employee attribute value. - list_employee_attribute_definitions (read-only): List the definitions of those custom attributes — their names and types. Read this to know what the values mean. - get_employee_attribute_definition (read-only): Get one attribute definition. - list_taxonomies (read-only): List the taxonomies — the classification systems the company applies to its people and posts. - get_taxonomy (read-only): Get one taxonomy. - list_taxonomy_labels (read-only): List the labels available inside the taxonomies. - get_taxonomy_label (read-only): Get one taxonomy label. - create_department (write, needs approval): Create a department. The only creation this connector performs on Lucca. - update_department (write, needs approval): Rename a department or move it under another one. Example prompts: - "List employees from the directory — from Lucca API V5." - "Get one employee's details by id — from Lucca API V5." - "List the organization's departments — from Lucca API V5." --- ## Magento URL: https://www.proxoria.com/en/connectors/magento Connect your AI to Magento. Your own Magento / Adobe Commerce store: browse orders, invoices, shipments, credit memos, customers, products and their attributes, categories and what is in them, stock by warehouse, cart price rules and coupons, CMS content and the store's reference tables (customer groups, countries, currencies, store views, tax rates); comment orders, adjust stock, and create or update products and CMS pages (disabled until a human publishes them). Connects to your instance's REST API with an integration access token. Authentication: connect Magento with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_orders (read-only): List sales orders, newest first. Filter by status, customer email or creation date. Returns a compact view; use get_order for the full detail of one order. - get_order (read-only): Get one order in full: line items, addresses, payment, totals and status history. - add_order_comment (write, needs approval): Add a comment to an order's history, optionally changing its status. The customer is NOT notified and the comment stays internal unless explicitly asked for. Writes to the store, so it goes through the confirmation guardrail when enabled. - list_invoices (read-only): List invoices, newest first, optionally those of one order. - list_shipments (read-only): List shipments, newest first, optionally those of one order. - list_creditmemos (read-only): List credit memos (refunds), newest first, optionally those of one order. - list_customers (read-only): Search customer accounts by email or name. - get_customer (read-only): Get one customer account in full, including addresses. - list_products (read-only): List catalog products, optionally filtered by name or SKU. - get_product (read-only): Get one product by SKU, with its attributes, media and prices. - create_product (write, needs approval): Create a simple product, DISABLED — it does not appear on the storefront until a human enables it in Magento. Writes to the store, so it goes through the confirmation guardrail when enabled. - update_product (write, needs approval): Update a product's name, price or weight. Deliberately cannot enable or disable it: putting a product live stays a human action in Magento. Writes to the store, so it goes through the confirmation guardrail when enabled. - list_categories (read-only): Get the category tree of the catalog. - get_stock (read-only): Get the stock item of one product: quantity and in-stock flag. - list_low_stock (read-only): List products whose quantity has fallen to or below a threshold — the restocking view. - update_stock_quantity (write, needs approval): Set the stock quantity of a product. Writes to the store, so it goes through the confirmation guardrail when enabled. - list_cms_pages (read-only): List CMS pages of the store, optionally filtered by title or URL key. - get_cms_page (read-only): Get one CMS page by id, with its full content. - create_cms_page (write, needs approval): Create a CMS page, INACTIVE — it is not served on the storefront until a human activates it in Magento. Writes to the store, so it goes through the confirmation guardrail when enabled. - update_cms_page (write, needs approval): Update a CMS page's title, content or SEO metadata. Deliberately cannot activate or deactivate it: publishing stays a human action in Magento. Writes to the store, so it goes through the confirmation guardrail when enabled. - list_cms_blocks (read-only): List CMS blocks (reusable content snippets), optionally filtered by title. - list_cart_price_rules (read-only): The cart price rules in force — the discounts a customer can get, with their dates, customer groups and websites. This is what explains a total that is not the sum of the line prices. - list_coupons (read-only): Coupon codes with the rule they belong to, how many times each may be used and how many times it already was. - list_category_products (read-only): The products assigned to a category, with their position — what the category page actually shows. - list_product_attributes (read-only): Product attributes with their code, type and — for dropdowns — their options. Needed before filtering or writing a custom attribute, whose code is instance-specific. - list_inventory_sources (read-only): The physical sources stock is held at (Magento's multi-source inventory): warehouses, shops, their address and whether each is enabled. - list_source_items (read-only): Stock per source: how much of a SKU sits at each warehouse or shop, and its status there. `get_stock` gives the aggregate; this says where it actually is. - list_reference_data (read-only): List one of the store's reference tables, to get the ids and codes the other tools report: customer groups, countries, currencies, store views, websites, product attribute sets, tax rates or tax classes. Example prompts: - "List sales orders, newest first. Filter by status, customer email or creation date. Returns a compact view; use get_order for the full detail of one order — from Magento." - "Get one order in full: line items, addresses, payment, totals and status history — from Magento." - "List invoices, newest first, optionally those of one order — from Magento." --- ## Mailchimp URL: https://www.proxoria.com/en/connectors/mailchimp Connect your AI to Mailchimp. Mailchimp Marketing: audiences and their contacts, segments, tags, merge fields and interests, campaigns drafted and given their content, classic automations, landing pages and Facebook ads, the connected store with its products, carts and orders, and the full reporting suite — per campaign, per link, per recipient. Nothing here sends an email. Authentication: connect Mailchimp with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_audiences (read-only): List audiences (lists) of the account. - list_members (read-only): List members (subscribers) of an audience. - add_member (write, needs approval): Add (or subscribe) a contact to an audience. Writes to Mailchimp, so it goes through the confirmation guardrail when enabled. - list_campaigns (read-only): List campaigns, optionally filtered by status. - get_campaign_report (read-only): Get the performance report of a campaign: opens, clicks, bounces, unsubscribes, plus `list_stats` (this audience's own averages) and `industry_stats` (the sector's) — the two figures that tell you whether a result is good, rather than merely what it is. - create_audience (write, needs approval): Create an audience. Mailchimp demands the details that make a mailing lawful, so they must come from the customer and cannot be invented: a physical postal address shown in every footer, the reminder telling recipients why they are hearing from you, and the default sender identity. Note that Mailchimp bills contacts per audience — the same person in two audiences counts twice. - add_members (write, needs approval): Add several contacts to an audience in one call. Each contact's `status` is a statement about consent: 'pending' asks them to confirm by email, 'subscribed' asserts they already agreed. It defaults to 'pending' — only say 'subscribed' when the customer confirms they hold that consent. - get_member (read-only): Get everything Mailchimp holds on one contact: status and unsubscribe reason, merge fields, tags, language, VIP flag, source, its own average open and click rates, a 1-to-5 rating, location — and the consent trail (signup and opt-in IP and timestamp, marketing permissions). Addressed by email; the contact id is derived here. - get_member_activity (read-only): Get a contact's recent activity — the campaigns they opened, clicked or bounced on. Narrow it with `activityFilters` to ask only for the events that matter, rather than reading the whole feed to find them. - search_members (read-only): Search contacts across every audience by name or email, without knowing which audience they belong to. - list_segments (read-only): List an audience's segments and tags — how the audience is actually cut up for targeting. - get_segment (read-only): Get one segment, with how many contacts it currently holds and — for a rule-based segment — the conditions behind it. Read an existing segment before writing one: its `options` show the exact shape a condition takes for each field type. - create_segment (write, needs approval): Create a segment of an audience, in one of two shapes. Give `emails` for a static segment — a fixed list you fill yourself, which is what a tag is. Give `match` and `conditions` for a rule-based segment, which Mailchimp recalculates on its own as contacts change. A condition's exact shape depends on the field it tests, so copy one from an existing segment read with get_segment rather than guessing. - update_segment (write, needs approval): Rename a segment or replace its conditions. Mailchimp requires the name on every call, even when only the conditions change — so pass the current name if you are not renaming it. Replacing the conditions replaces them entirely. - update_segment_members (write, needs approval): Add or remove contacts of a static segment, up to 500 of each per call. Only works on a static segment: a rule-based one has no fixed membership to edit, change its conditions with update_segment instead. Addresses absent from the audience are ignored rather than created. - search_tags (read-only): Search an audience's tags by name, or list them all when no name is given. Call this before add_member_tags: Mailchimp creates a tag silently when the name does not match an existing one, so a difference of case or spelling quietly makes a second tag nobody notices. - add_member_tags (write, needs approval): Tag or untag a contact. Each tag carries its own status, so one call can both add and remove. Tags are the everyday way to segment in Mailchimp — a contact added without one is hard to reach again. - create_campaign (write, needs approval): Create a campaign. It is born as a draft and stays one: this connector exposes no way to send or schedule, because that reaches thousands of people and cannot be undone. Give it content with set_campaign_content, then a human sends it from Mailchimp. - update_campaign (write, needs approval): Change a draft campaign's subject, sender, preview text or recipients, without touching its content. Only a campaign that has not been sent can be changed. - get_campaign_content (read-only): Get a campaign's content, in HTML and plain text. - set_campaign_content (write, needs approval): Set a campaign's content — raw HTML, plain text, or a template with its sections filled in. Replaces whatever was there. The campaign remains a draft until a human sends it. - list_templates (read-only): List the email templates of the account. - get_template (read-only): Get one template — its name, category and when it was last edited. - create_template (write, needs approval): Create a template from raw HTML. Mailchimp's API only creates Classic templates: the result will not open in the drag-and-drop editor, so the customer will have to come back here — or to raw HTML — to change it. Mark the parts a campaign should fill with mc:edit attributes, otherwise the template has no editable sections. - update_template (write, needs approval): Replace a template's name and HTML. Despite being a PATCH, Mailchimp requires both every time — there is no way to amend part of a template, so read it first and send back the whole thing. Classic templates only, like create_template. - list_campaign_reports (read-only): List the reports of sent campaigns. Bound the period with `sinceSendTime` and `beforeSendTime` to compare one quarter against another — Mailchimp has no comparison endpoint, so this is how a trend gets built. - get_campaign_open_details (read-only): List who opened a campaign, contact by contact — the report to cross with a CRM to find who is paying attention. Automated opens are excluded by default; since mail apps began preloading images, counting them overstates interest. - get_campaign_click_details (read-only): List the links of a campaign and how many contacts clicked each one. - get_campaign_email_activity (read-only): Replay every recorded action of a campaign contact by contact — opens, clicks and bounces, each with its timestamp. The heaviest report of the connector: prefer get_campaign_open_details or get_campaign_click_details for a plain answer, and use this with `since` to pick up only what has happened since a given moment. - get_campaign_recipients (read-only): List who a campaign was sent to, with each contact's delivery status. - get_campaign_abuse_reports (read-only): List the spam complaints a campaign drew. Small, and the earliest warning that deliverability is slipping. - get_audience (read-only): Get one audience: its name, the postal address and permission reminder Mailchimp shows in every footer, the default sender identity, and its stats — member count, unsubscribes, average open and click rate, and when it last sent. - update_audience (write, needs approval): Change an audience's settings. The contact block and the permission reminder are what make a mailing lawful and appear in every footer, so they come from the customer and are never invented. Only the fields you give are touched. - list_audience_abuse_reports (read-only): List the spam complaints raised against this audience, across all its campaigns. A rising count here is the earliest sign that deliverability is about to suffer. - get_audience_abuse_report (read-only): Get one spam complaint: who reported it, on which campaign, and when. - get_audience_activity (read-only): The audience's recent activity, day by day: sends, opens, clicks, subscribes, unsubscribes and bounces. The shape of the list's health over time rather than one campaign's numbers. - list_audience_email_clients (read-only): Which email clients this audience actually reads in, with a count per client. Worth checking before deciding how much a rendering quirk matters. - list_member_tags (read-only): List the tags on one contact. Give the email address; Mailchimp's own identifier is derived from it. - list_segment_members (read-only): List the contacts a segment currently contains. For a rule-based segment this is evaluated when you ask, so it can differ from the last time you looked. - add_segment_member (write, needs approval): Add one contact to a static segment. Rule-based segments decide their own membership and reject this. - get_campaign (read-only): Get one campaign: its type, status, audience or segment, subject, sender, schedule, and the counts it has accumulated. Read this before changing anything — status tells you whether it is still a draft. - list_campaign_feedback (read-only): List the internal review comments left on a campaign by the team, before it went out. Not customer feedback — this is the draft's comment thread. - get_campaign_feedback (read-only): Get one review comment on a campaign, with who wrote it and which block it points at. - get_campaign_send_checklist (read-only): Mailchimp's own pre-flight check on a draft: what is still missing, what is only a warning, and whether the campaign is ready to send. The right thing to read before telling anyone a draft is finished. - search_campaigns (read-only): Search campaigns by name, subject or content across the account. The way to find a campaign when all you have is a phrase from it. - list_merge_fields (read-only): List an audience's merge fields — the custom fields a contact can carry, with their tag, type and whether they are required. Read this before adding contacts: the tags are what `mergeFields` expects, and an invented one is silently dropped. - get_merge_field (read-only): Get one merge field with its type, default value and options. - list_interest_categories (read-only): List an audience's interest categories — the groups a contact can opt into, as they appear on the signup form. - get_interest_category (read-only): Get one interest category, with how it is presented on the form. - list_interests (read-only): List the interests inside a category, with the number of subscribers who picked each. Their ids are what a segment condition on interests refers to. - get_interest (read-only): Get one interest and how many subscribers it holds. - list_member_notes (read-only): List the notes left on a contact by the team. Free text that decides nothing, and often where the context behind a contact's status actually lives. - get_member_note (read-only): Get one note on a contact, with its author and date. - add_member_note (write, needs approval): Add a note to a contact. It is internal, visible to the team in Mailchimp, and never sent to the contact. - list_automations (read-only): List the classic automations of the account — the sequences that fire on their own, such as a welcome series. Each carries its status, the audience it runs on and its aggregate report. - get_automation (read-only): Get one automation: its trigger, its status, the audience or segment it targets, and how it has performed since it started. - list_automation_emails (read-only): List the emails that make up an automation, in order, each with its delay and its own figures. This is how you see which step of a sequence loses people. - get_automation_email (read-only): Get one email of an automation, with its delay, its content settings and its report. - list_automation_email_queue (read-only): List the contacts currently waiting for one automation email — who is mid-sequence, and when each is due to receive it. - get_automation_email_subscriber (read-only): Where one contact stands in an automation email's queue. Give the email address; Mailchimp's identifier is derived from it. - list_automation_removed_subscribers (read-only): List the contacts taken out of an automation before it finished. The complement of the queue: who dropped out of the sequence. - get_automation_removed_subscriber (read-only): Check whether one contact was removed from an automation, and when. - list_campaign_folders (read-only): List the folders campaigns are filed in, with how many each holds. The way to find a campaign by where it was put rather than by its name. - get_campaign_folder (read-only): Get one campaign folder and its campaign count. - list_template_folders (read-only): List the folders templates are filed in, with how many each holds. - get_template_folder (read-only): Get one template folder and its template count. - list_marketing_audiences (read-only): List audiences through Mailchimp's newer contact-oriented endpoints. These are the same audiences `list_audiences` returns; what differs is the shape of the response and the fact that contacts here are addressed by their own id rather than by a hash of their email. - get_marketing_audience (read-only): Get one audience through the newer endpoints. The id is the same as an audience's list id. - list_audience_contacts (read-only): List an audience's contacts through the newer endpoints, each with its own contact id. Use `list_members` instead when you need the merge fields, tags and consent history — that response is richer. - get_audience_contact (read-only): Get one contact by its contact id, which is not the hash the member endpoints use. Take it from list_audience_contacts. - list_connected_sites (read-only): List the websites connected to the account, each with its domain and whether its tracking pixel is installed and firing. - get_connected_site (read-only): Get one connected site, with its store link and the state of its script. - list_facebook_ads (read-only): List the Facebook and Instagram ads run from Mailchimp, with their status, budget, audience and results. - get_facebook_ad (read-only): Get one Facebook or Instagram ad: its creative, its targeting, its spend and what it returned. - get_campaign_abuse_report (read-only): Get one spam complaint against a campaign: who raised it and when. - get_campaign_advice (read-only): Mailchimp's own read on a campaign's numbers — where it beat or missed the account's and the industry's averages, in plain sentences. Useful as a second opinion, not as a verdict. - get_campaign_link_clicks (read-only): The figures for one link in a campaign: total clicks, unique clicks, and its share of all clicks. The link id comes from get_campaign_click_details. - list_link_clickers (read-only): Who clicked one particular link, and how many times each. This is the report that turns a click rate into a list of people to follow up. - get_link_clicker (read-only): Whether one contact clicked a particular link, and how often. - get_campaign_domain_performance (read-only): How a campaign performed per recipient domain — gmail.com against outlook.com and the rest, with opens, clicks and bounces for each. Where a deliverability problem shows itself, because it almost always affects one mailbox provider and not the others. - get_campaign_ecommerce_activity (read-only): The products bought after a campaign, with revenue per product. Only meaningful when a store is connected and its orders are synced. - get_campaign_eepurl (read-only): How the campaign's public archive page and its social shares performed: views, likes, and referrers. - get_member_email_activity (read-only): One contact's timeline for one campaign: every open, every click and every bounce, timestamped. The finest grain the API offers on a single person. - get_campaign_locations (read-only): Where a campaign was opened, by country and region, with a count for each. - get_member_open_details (read-only): Whether one contact opened a campaign, how many times and when. Bear in mind that image preloading inflates opens, so a single machine open proves little. - get_campaign_recipient (read-only): Whether one contact was sent a campaign, and what became of it — delivered, bounced, or held. - list_campaign_sub_reports (read-only): The individual reports of a multi-variant campaign — an A/B test or an RSS series — each with its own figures. Reading only the parent report of such a campaign averages the variants together and hides which one won. - list_campaign_unsubscribes (read-only): Who unsubscribed because of this campaign, with the reason each gave. The most direct feedback a campaign produces, and the least read. - get_campaign_unsubscribe (read-only): The unsubscribe of one contact on one campaign, with its reason and timestamp. - list_facebook_ad_reports (read-only): The report for every Facebook and Instagram ad: spend, reach, clicks and the revenue Mailchimp attributes to it. - get_facebook_ad_report (read-only): The report for one ad, with its audience and what it returned. - get_facebook_ad_ecommerce_activity (read-only): The products bought after one ad, with revenue per product. - list_landing_page_reports (read-only): The report for every landing page: visits, unique visitors, subscribes and conversion rate. - get_landing_page_report (read-only): The report for one landing page, with its traffic and what it converted. - list_survey_reports (read-only): The report for every survey: how many were started, how many finished, and the drop-off between. - get_survey_report (read-only): The report for one survey, with its completion figures. - list_survey_question_reports (read-only): The per-question report of a survey: how many answered each, and where people gave up. - get_survey_question_report (read-only): The report for one survey question. - list_survey_question_answers (read-only): The individual answers to one survey question. For a free-text question this is the raw material worth summarising rather than counting. - list_survey_responses (read-only): The responses to a survey, one per respondent, with who submitted each where the survey identified them. - get_survey_response (read-only): One survey response in full, question by question. - list_all_orders (read-only): Every order across every connected store, newest first. The account-wide view, when you do not want to walk store by store. - list_stores (read-only): List the connected stores. A store is what Mailchimp attributes orders and revenue to, and it is tied to one audience. - create_store (write, needs approval): Connect a store to Mailchimp. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id`, `list_id`, `name` and `currency_code`. - get_store (read-only): Get one store by id. - update_store (write, needs approval): Change a store. Only the fields you send are touched. - delete_store (write, needs approval): Delete a store from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_carts (read-only): List the carts of a store. Carts include the abandoned ones, which is what an abandoned-cart automation acts on. - create_cart (write, needs approval): Add a cart to a store. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id`, `currency_code`, `customer`, `order_total` and `lines`. - get_cart (read-only): Get one cart by id. - update_cart (write, needs approval): Change a cart. Only the fields you send are touched. - delete_cart (write, needs approval): Delete a cart from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_cart_lines (read-only): List the lines of a cart. - add_cart_line (write, needs approval): Add a cart line to a cart. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id`, `product_id`, `product_variant_id`, `quantity` and `price`. - get_cart_line (read-only): Get one cart line by id. - update_cart_line (write, needs approval): Change a cart line. Only the fields you send are touched. - delete_cart_line (write, needs approval): Delete a cart line from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_store_customers (read-only): List the customers of a store. A store customer is not an audience contact; Mailchimp links the two by email address. - create_store_customer (write, needs approval): Add a store customer to a store. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id` and `opt_in_status`. - get_store_customer (read-only): Get one store customer by id. - update_store_customer (write, needs approval): Change a store customer. Only the fields you send are touched. - upsert_store_customer (write, needs approval): Create a store customer, or replace it entirely if its id already exists. Unlike an update, anything you leave out is cleared. Mailchimp requires `id`. - delete_store_customer (write, needs approval): Delete a store customer from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_store_orders (read-only): List the orders of a store. Orders are what turn a campaign's opens and clicks into revenue in the reports. - create_order (write, needs approval): Add a order to a store. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id`, `customer`, `currency_code`, `order_total` and `lines`. - get_order (read-only): Get one order by id. - update_order (write, needs approval): Change a order. Only the fields you send are touched. - upsert_order (write, needs approval): Create a order, or replace it entirely if its id already exists. Unlike an update, anything you leave out is cleared. Mailchimp requires `id`. - delete_order (write, needs approval): Delete a order from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_order_lines (read-only): List the lines of an order. - add_order_line (write, needs approval): Add a order line to an order. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id`, `product_id`, `product_variant_id`, `quantity` and `price`. - get_order_line (read-only): Get one order line by id. - update_order_line (write, needs approval): Change a order line. Only the fields you send are touched. - delete_order_line (write, needs approval): Delete a order line from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_products (read-only): List the products of a store. The catalogue Mailchimp draws on for product blocks and recommendations. - create_product (write, needs approval): Add a product to a store. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id`, `title` and `variants`. - get_product (read-only): Get one product by id. - update_product (write, needs approval): Change a product. Only the fields you send are touched. - upsert_product (write, needs approval): Create a product, or replace it entirely if its id already exists. Unlike an update, anything you leave out is cleared. Mailchimp requires `id`. - delete_product (write, needs approval): Delete a product from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_product_images (read-only): List the images of a product. - add_product_image (write, needs approval): Add a product image to a product. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id` and `url`. - get_product_image (read-only): Get one product image by id. - update_product_image (write, needs approval): Change a product image. Only the fields you send are touched. - delete_product_image (write, needs approval): Delete a product image from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_product_variants (read-only): List the variants of a product. A variant is a size, a colour, a format. - create_product_variant (write, needs approval): Add a product variant to a product. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id` and `title`. - get_product_variant (read-only): Get one product variant by id. - update_product_variant (write, needs approval): Change a product variant. Only the fields you send are touched. - upsert_product_variant (write, needs approval): Create a product variant, or replace it entirely if its id already exists. Unlike an update, anything you leave out is cleared. Mailchimp requires `id` and `title`. - delete_product_variant (write, needs approval): Delete a product variant from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_promo_rules (read-only): List the promotion rules of a store. The rule holds the discount; the codes under it are what customers type. - create_promo_rule (write, needs approval): Add a promotion rule to a store. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id`, `description`, `amount`, `type` and `target`. - get_promo_rule (read-only): Get one promotion rule by id. - update_promo_rule (write, needs approval): Change a promotion rule. Only the fields you send are touched. - delete_promo_rule (write, needs approval): Delete a promotion rule from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. - list_promo_codes (read-only): List the promo codes of a rule. - create_promo_code (write, needs approval): Add a promo code under a rule. The id is yours to choose and is how your own system refers to it afterwards. Mailchimp requires `id`, `code` and `redemption_url`. - get_promo_code (read-only): Get one promo code by id. - update_promo_code (write, needs approval): Change a promo code. Only the fields you send are touched. - delete_promo_code (write, needs approval): Delete a promo code from Mailchimp. It takes the figures that hang off it with it, and cannot be undone. Example prompts: - "List audiences (lists) of the account — from Mailchimp." - "List members (subscribers) of an audience — from Mailchimp." - "List campaigns, optionally filtered by status — from Mailchimp." --- ## Mautic URL: https://www.proxoria.com/en/connectors/mautic Connect your AI to Mautic. Your own Mautic marketing automation instance: browse contacts with their activity and notes, segments and their filters, campaigns, emails, companies, forms, landing pages, tags, point rules, webhooks and saved reports; create and update contacts and companies, adjust a score, move contacts in and out of segments and campaigns, record notes, set or lift a Do Not Contact flag, and draft emails. Nothing here sends. Connects to your instance's REST API with Mautic credentials. Authentication: connect Mautic with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_contacts (read-only): List or search contacts. `search` matches names and emails, and also accepts Mautic search commands (email:acme.com, segment:newsletter, is:anonymous). - get_contact (read-only): Get one contact with all of its fields, tags, segments and scoring points. - get_contact_activity (read-only): Timeline of one contact: page views, email opens and clicks, form submissions, campaign steps. The fastest way to answer 'what has this lead actually done?'. - create_contact (write, needs approval): Create a contact. Writes to Mautic, so it goes through the confirmation guardrail when enabled. - update_contact (write, needs approval): Update fields of an existing contact. Only the fields passed are changed; the rest are left alone. - adjust_contact_points (write, needs approval): Add or subtract scoring points on a contact. Use a negative number to subtract. - list_segments (read-only): List contact segments with their name, alias and description. - add_contact_to_segment (write, needs approval): Add a contact to a segment. Writes to Mautic. - remove_contact_from_segment (write, needs approval): Remove a contact from a segment. Writes to Mautic. - list_campaigns (read-only): List campaigns with their name, category and published state. - get_campaign (read-only): Get one campaign with its events and the segments that feed it. - add_contact_to_campaign (write, needs approval): Add a contact to a campaign — the contact then starts flowing through its steps. Writes to Mautic. - remove_contact_from_campaign (write, needs approval): Remove a contact from a campaign. Writes to Mautic. - list_emails (read-only): List emails with their name, subject, type and published state. - get_email (read-only): Get one email with its subject, HTML body and settings. - create_email (write, needs approval): Draft an email. Created UNPUBLISHED — a human publishes and sends it from Mautic. A list email must name its target segments via segmentIds. Use {contactfield=firstname} style tokens for personalisation. - update_email (write, needs approval): Update an email's name, subject or HTML body. Cannot publish or send it — that stays a human action in Mautic. - list_companies (read-only): List companies with their name, industry and location. - get_company (read-only): Get one company with all of its fields and the contacts attached to it. - create_company (write, needs approval): Create a company record. Writes to Mautic, so it goes through the confirmation guardrail when enabled. - get_segment (read-only): Get one segment with its filters — the rules that decide who is in it — and its counts. - list_contact_notes (read-only): List the notes recorded on a contact — what people wrote about them, with dates. - add_contact_note (write, needs approval): Add a note to a contact — an internal record, never sent to them. Writes to Mautic, so it goes through the confirmation guardrail when enabled. - set_do_not_contact (write, needs approval): Add a contact to, or remove them from, the Do Not Contact list for a channel. This is the consent switch: adding stops Mautic sending on that channel. Writes to Mautic, so it goes through the confirmation guardrail when enabled. - list_records (read-only): List one of Mautic's other objects: forms, landing pages, assets, tags, stages, categories, point actions and triggers, webhooks, users or saved reports. Ask for the entity by name. Example prompts: - "List or search contacts. `search` matches names and emails, and also accepts Mautic search commands (email:acme.com, segment:newsletter, is:anonymous) — from Mautic." - "Get one contact with all of its fields, tags, segments and scoring points — from Mautic." - "Timeline of one contact: page views, email opens and clicks, form submissions, campaign steps. The fastest way to answer 'what has this lead actually done?' — from Mautic." --- ## Nextcloud URL: https://www.proxoria.com/en/connectors/nextcloud Connect your AI to Nextcloud. Your own Nextcloud: browse and search folders and files, read text files, create folders, upload text files, copy, move, rename and delete items, read a file's versions and the trash, follow recent activity, and manage shares (public links, users and groups) including their permissions, expiry and password. Connects to your instance with your login name and an app password. Authentication: connect Nextcloud with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_files (read-only): List the folders and files directly inside a folder (one level deep), with type, size and last-modified time. Use the returned path to descend into a subfolder or to act on an item. - read_text_file (read-only): Read the contents of a text file (notes, markdown, code, CSV…). Binary files are not supported; very large files are truncated. - create_folder (write, needs approval): Create a folder at the given path (the parent folder must already exist). Writes to Nextcloud, so it goes through the confirmation guardrail when enabled. - upload_text_file (write, needs approval): Create or overwrite a text file with the given content (the parent folder must already exist). Writes to Nextcloud, so it goes through the confirmation guardrail when enabled. - delete_item (write, needs approval): Delete a file or folder (a folder is removed with everything inside it). Writes to Nextcloud, so it goes through the confirmation guardrail when enabled. - move_item (write, needs approval): Move or rename a file or folder to a new path (the destination's parent folder must already exist). Writes to Nextcloud, so it goes through the confirmation guardrail when enabled. - list_shares (read-only): List existing shares (public links and shares with users or groups), optionally only those on a given file or folder path. - create_share (write, needs approval): Share a file or folder. Choose 'public_link' for a shareable URL (optionally password protected and read-only), or 'user'/'group' to share with someone on the instance. Writes to Nextcloud, so it goes through the confirmation guardrail when enabled. - delete_share (write, needs approval): Remove an existing share by its id (from list_shares). Writes to Nextcloud, so it goes through the confirmation guardrail when enabled. - get_user_info (read-only): Get the connected user's account details: display name, email, groups and storage quota (used, total, free). - update_share (write, needs approval): Change an existing share: make it read-only or writable again, set or clear its expiry date, or put a password on a public link. Writes to Nextcloud, so it goes through the confirmation guardrail when enabled. - search_files (read-only): Search the account's files by name through Nextcloud's own search, when you do not know which folder holds what you are after. - list_activity (read-only): Recent activity on the account's files: what was created, changed, shared or deleted, by whom and when. Requires the Activity app (shipped and enabled by default). - list_users_and_groups (read-only): List the instance's user accounts or its groups. Nextcloud restricts this to administrators, so it answers 403 for an ordinary account — that is a permission answer, not a broken connection. - copy_item (write, needs approval): Copy a file or folder to another path, leaving the original in place. Writes to Nextcloud, so it goes through the confirmation guardrail when enabled. - list_trash (read-only): List what is in the account's trash, with the path each item came from and when it was deleted — the first step before restoring something. - list_file_versions (read-only): List the stored versions of a file — when each was saved and how big it was — using the file id that list_files reports. Example prompts: - "List the folders and files directly inside a folder (one level deep), with type, size and last-modified time. Use the returned path to descend into a subfolder or to act on an item — from Nextcloud." - "Read the contents of a text file (notes, markdown, code, CSV…). Binary files are not supported; very large files are truncated — from Nextcloud." - "List existing shares (public links and shares with users or groups), optionally only those on a given file or folder path — from Nextcloud." --- ## noCRM.io URL: https://www.proxoria.com/en/connectors/nocrm Connect your AI to noCRM.io. noCRM.io: search and read your sales leads with their status, step, amount and comments; create and update leads, move them along the pipeline, mark them won or lost, and post comments. Also lists pipelines, steps, client folders, prospecting lists and users. Connects with your own noCRM login, so the AI sees exactly what you see. Authentication: connect noCRM.io with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_leads (read-only): List leads, newest first by default. Filter by status (todo, standby, won, cancelled, lost), by pipeline step, by owner, by tags, or to starred leads only. Returns the lead summary; call get_lead for the full record. - get_lead (read-only): Get one lead in full: description, contact details, amount, probability, step, tags, reminders and related user, team and client folder. - create_lead (write, needs approval): Create a lead. It is owned by the connected user — noCRM does not allow assigning it to someone else with a personal login. Write action: subject to the confirmation guardrail when it is enabled. - update_lead (write, needs approval): Update a lead: move it to another step, change its status (marking it won, lost or cancelled closes it), adjust amount, probability, tags, reminder or closing date. Write action: subject to the confirmation guardrail when it is enabled. - list_lead_comments (read-only): List the comments posted on a lead, oldest first. - add_lead_comment (write, needs approval): Post a comment on a lead — the usual way to log a call, a meeting outcome or a next step. Write action: subject to the confirmation guardrail when it is enabled. - list_steps (read-only): List the pipeline steps a lead can sit in, in order. Use it before moving a lead so the step name is one the account actually has. - list_pipelines (read-only): List the sales pipelines of the account. - list_client_folders (read-only): List client folders — the customer records leads can be attached to once they are won. - list_prospecting_lists (read-only): List the prospecting lists (the spreadsheets sales reps work through before a prospect becomes a lead). - list_users (read-only): List the users of the account, with their id — needed to filter leads by owner. - delete_lead (write, needs approval): Delete a lead. It goes with its comments, its attachments and its history, and noCRM offers no way back — read it first if there is any doubt. - delete_leads (write, needs approval): Delete several leads at once. Same finality as delete_lead, multiplied — this is the single most destructive call the connector exposes. - add_lead_to_client (write, needs approval): File a lead under a client folder, which is how noCRM groups everything belonging to one customer. Give the folder id, or a name to create the folder on the way. - assign_lead (write, needs approval): Hand a lead to a salesperson. Give `userId` for a named person, or a `mode` to let noCRM pick: random, round_robin, or random_in_a_team with `teamId`. - duplicate_lead (write, needs approval): Copy a lead into a new one, which is how a returning customer gets a second deal without losing the first. - send_lead_email_from_template (write, needs approval): Send an email to the lead's contact from one of the account's templates. This really leaves the building and cannot be recalled: the customer receives it. - create_lead_follow_ups (write, needs approval): Apply a follow-up template to a lead, which creates the whole sequence of tasks it describes. - list_unassigned_leads (read-only): List the leads nobody owns. The question a sales manager actually asks, and the one the general lead list answers badly. - list_lead_duplicates (read-only): The leads noCRM considers duplicates of this one, by contact details. Worth checking before creating anything, because a duplicated lead splits a customer's history in two. - list_lead_history (read-only): Everything that happened to a lead, in order: step changes, reassignments, status changes, with who did each and when. The record of how a deal actually went. - list_lead_attachments (read-only): List the files attached to a lead — names, sizes and types. The contents are not returned. - get_lead_attachment (read-only): Get the metadata of one attachment, including the link noCRM serves it from. - get_lead_business_card (read-only): The lead's contact details on their own — name, company, phone, email — without the rest of the record. - update_lead_comment (write, needs approval): Rewrite a comment on a lead. - delete_lead_comment (write, needs approval): Delete a comment from a lead. - list_follow_ups (read-only): List the follow-up sequences of the account, with their tasks and due dates. - update_follow_up_task (write, needs approval): Change one task of a follow-up — mark it done, move its date, rewrite it. - create_client_folder (write, needs approval): Create a client folder — the box that gathers every lead belonging to one customer, with the billing and shipping details. - get_client_folder (read-only): Get one client folder with its details and the leads filed under it. - update_client_folder (write, needs approval): Change a client folder's details. Only the fields you send are touched. - delete_client_folder (write, needs approval): Delete a client folder. The leads inside it are not deleted, they simply lose the folder. - list_teams (read-only): List the sales teams of the account, with their members. - get_team (read-only): Get one team and who is in it. - create_team (write, needs approval): Create a sales team. - update_team (write, needs approval): Rename a team. - delete_team (write, needs approval): Delete a team. Its members keep their accounts and their leads. - add_team_member (write, needs approval): Add a user to a team. - remove_team_member (write, needs approval): Take a user out of a team. - get_user (read-only): Get one user: their name, email, role and whether the account is active. - get_step (read-only): Get one pipeline step by id, so a step id coming back on a lead reads as the stage it is. - list_categories (read-only): List the lead categories of the account. - list_predefined_tags (read-only): List the tags the account has defined in advance. Reading these first keeps an assistant from inventing a tag that nobody else uses. - list_fields (read-only): List the custom fields of the account, with their type and their possible values. Required before writing one: the names here are what create_lead and update_lead expect, and an invented one is ignored. - list_activities (read-only): List the activity types the account uses — call, meeting, email — the vocabulary a logged action has to pick from. - list_prospects (read-only): List the prospects across the account's prospecting lists — the rows that have not become leads yet. Filter by list, by owner or by status to narrow it before paging, because the account's daily request budget is shared with everything else. - list_prospects_called_from (read-only): Find the prospect a phone number belongs to. This is what turns an incoming call into a name before anyone picks up. - get_prospecting_list (read-only): Get one prospecting list with its columns and its counts. - create_prospecting_list (write, needs approval): Create a prospecting list — the sheet a rep works through before anything becomes a lead. - assign_prospecting_list (write, needs approval): Hand a prospecting list to a salesperson. - comment_prospecting_list (write, needs approval): Leave a comment on a prospecting list. - add_prospect (write, needs approval): Add a prospect to a prospecting list. - update_prospect (write, needs approval): Change a prospect's own attributes — its status, its owner, its comment. - update_prospect_fields (write, needs approval): Change the column values of a prospect — the data of the sheet itself, as opposed to its status and ownership. - create_lead_from_prospect (write, needs approval): Turn a prospect into a lead, which is the moment a name in a sheet enters the pipeline. The row keeps its place and gains a link to the lead. - delete_prospect (write, needs approval): Delete a prospect from a prospecting list. - comment_prospect (write, needs approval): Leave a comment on a prospect — what was said when nobody picked up. - update_prospect_comment (write, needs approval): Rewrite a comment on a prospect. Example prompts: - "List leads, newest first by default. Filter by status (todo, standby, won, cancelled, lost), by pipeline step, by owner, by tags, or to starred leads only. Returns the lead summary; call get_lead for the full record — from noCRM.io." - "Get one lead in full: description, contact details, amount, probability, step, tags, reminders and related user, team and client folder — from noCRM.io." - "List the comments posted on a lead, oldest first — from noCRM.io." --- ## Odoo URL: https://www.proxoria.com/en/connectors/odoo Connect your AI to Odoo. Your own Odoo 19+ (Community or Enterprise): CRM leads and contacts, sales and purchase orders, invoices, payments and the general ledger, products, stock and transfers, manufacturing orders, projects, tasks and timesheets, employees, time off and expenses; create leads, contacts, quotations and tasks, confirm an order and log internal notes. Connects to your instance's JSON-2 API with a per-user API key. For Odoo 18 and earlier, use the Odoo (RPC) connector. Authentication: connect Odoo with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_leads (read-only): List CRM leads and opportunities, optionally filtered by a search on name, contact or email. - get_lead (read-only): Get one lead/opportunity in detail, including its description. - create_lead (write, needs approval): Create a CRM lead. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - update_lead (write, needs approval): Update a lead/opportunity: move it to another stage (ids from list_reference_data with kind=crm_stages), revise the expected revenue or probability, or correct its contact details. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - log_note (write, needs approval): Log an internal note on a lead, a contact or a sale order (it lands in the record's chatter, visible to the team, not emailed to the customer). Writes to the ERP, so it goes through the confirmation guardrail when enabled. - list_partners (read-only): List contacts and companies, optionally filtered by name/email or restricted to customers or suppliers. - get_partner (read-only): Get one contact/company in detail: address, VAT, website, internal notes. - create_partner (write, needs approval): Create a contact or company. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - update_partner (write, needs approval): Update a contact or company: address, email, phone, VAT number. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - list_sale_orders (read-only): List sale orders and quotations, optionally by status (draft and sent are quotations, sale is confirmed) or restricted to one customer. - get_sale_order (read-only): Get one sale order with its lines (product, quantity, unit price, subtotal). - create_quotation (write, needs approval): Create a sales quotation (a draft sale order) for a customer, with its product lines. It stays a quotation until someone confirms it — confirm_sale_order does that as a separate, explicit step. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - confirm_sale_order (write, needs approval): Confirm a quotation into a sales order — this is what triggers delivery and invoicing downstream, so it is a real commercial commitment. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - list_purchase_orders (read-only): List purchase orders (what the business buys), optionally by status or supplier. - get_purchase_order (read-only): Get one purchase order with its lines (product, quantity ordered, unit price, subtotal). - list_invoices (read-only): List invoices, customer ones by default — supplier bills on request — optionally by payment state (not_paid is the chasing view) or restricted to one partner. - get_invoice (read-only): Get one invoice with its lines and payment situation. - list_payments (read-only): List customer and supplier payments: amount, date, partner, journal and status. Use it to answer 'has this been paid, and when'. - list_journal_items (read-only): List accounting entries line by line — the general ledger. Filter by account (ids from list_reference_data with kind=accounts), by journal, by partner or by date range. This is the tool for 'what is on account 411 this quarter'. - list_products (read-only): List products with their price and available stock, optionally filtered by name or internal reference. - get_product (read-only): Get one product in detail: price, cost, stock, sales description. - list_transfers (read-only): List stock transfers — deliveries, receipts and internal moves — with their scheduled date and status. `late` is the shortcut for 'what should have shipped by now'. Requires the Inventory app. - get_transfer (read-only): Get one stock transfer with its lines: which product, how much was asked for and how much is actually reserved or done. - list_stock (read-only): List on-hand stock per product and location, with what is reserved and what is really available. Requires the Inventory app. - list_manufacturing_orders (read-only): List manufacturing orders: what is being produced, how much, when and in which state. Requires the Manufacturing app. - list_projects (read-only): List projects (requires the Project app on the instance). - list_tasks (read-only): List project tasks, optionally those of one project (requires the Project app). - get_task (read-only): Get one task in detail: description, assignees, deadline, and the hours allocated versus the hours actually logged against it. - create_task (write, needs approval): Create a task in a project, optionally assigned and with a deadline. Writes to the ERP, so it goes through the confirmation guardrail when enabled. - list_timesheets (read-only): List timesheet entries — who logged how many hours, on which project or task, when. Requires the Timesheets app. - list_employees (read-only): List employees with their job title, department and work contact details. Requires the Employees app. - list_time_off (read-only): List time-off requests: who is away, when, under which leave type, and whether it is approved. Requires the Time Off app. - list_expenses (read-only): List employee expenses with their amount and approval status. Requires the Expenses app. - list_tickets (read-only): List helpdesk tickets (requires the Helpdesk app — Odoo Enterprise only). - list_reference_data (read-only): List one of the instance's reference tables, to get the ids the other tools need: CRM stages, task stages, accounting journals, ledger accounts, warehouses, stock locations, departments, product categories, companies or users. - list_other_records (read-only): List one of the smaller business objects this Odoo may hold: job applicants, events and their registrations, marketing mailings, fleet vehicles, maintenance requests and equipment, repair orders or surveys. Each needs its Odoo app installed; ask for the entity by name, not by model. Example prompts: - "List CRM leads and opportunities, optionally filtered by a search on name, contact or email — from Odoo." - "Get one lead/opportunity in detail, including its description — from Odoo." - "List contacts and companies, optionally filtered by name/email or restricted to customers or suppliers — from Odoo." --- ## OpenProject URL: https://www.proxoria.com/en/connectors/openproject Connect your AI to OpenProject. Your own OpenProject: browse work packages with their comments, relations, attachments and watchers, projects and who can be assigned on them, versions, categories, news, saved views, logged time and the reference data; create and update work packages, comment them, relate them, add watchers and log time. Connects to your instance's API v3 with an OpenProject API token. Authentication: connect OpenProject with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_work_packages (read-only): List work packages, optionally within one project. `statusFilter` open/closed is the usual cut; ids for the other filters come from list_statuses, list_types, list_priorities and list_project_members. - get_work_package (read-only): Get one work package. Returns its `lockVersion` — the number an update needs — alongside its fields, with links resolved to {id, name} rather than raw HAL hrefs. - create_work_package (write, needs approval): Create a work package in a project. `typeId` comes from list_types, `priorityId` from list_priorities, `assigneeId` from list_project_members. Writes to OpenProject, so it goes through the confirmation guardrail when enabled. - update_work_package (write, needs approval): Update a work package's fields. Only what you pass changes. To add a comment, use add_work_package_comment instead. - list_work_package_activities (read-only): The activity trail of one work package: comments and the field changes that were made, oldest first. - add_work_package_comment (write, needs approval): Add a comment to a work package, changing nothing else. Writes to OpenProject. - list_work_package_relations (read-only): Relations of one work package — what it blocks, what blocks it, what it precedes or relates to. The way to answer 'what is holding this up?'. - create_work_package_relation (write, needs approval): Relate two work packages. Writes to OpenProject. - list_projects (read-only): List the projects visible to this token, with their id, name and identifier. - get_project (read-only): Get one project with its description and status. - list_project_members (read-only): Who can be assigned work on a project, with their user id and name. This is how a person's name becomes the id assigneeId expects — without touching the instance's global user directory. - list_statuses (read-only): Work package statuses with their ids, and which count as closed. - list_types (read-only): Work package types with their ids (Task, Milestone, Feature…). A type must be enabled on the target project. - list_priorities (read-only): Priorities with their ids (Low, Normal, High, Immediate). - list_time_entry_activities (read-only): Activities time can be logged against, with their ids. Read them from the time-entry form, which is where OpenProject exposes the values a given work package accepts. - list_time_entries (read-only): Logged time, optionally restricted to one project or one work package. - log_time (write, needs approval): Log time against a work package. `activityId` comes from list_time_entry_activities. Writes to OpenProject. - list_versions (read-only): Versions (milestones/sprints) of a project, with their status and dates — what a work package's `version` points at. - list_categories (read-only): Work package categories defined on a project, with their default assignee. - list_news (read-only): News posted on the instance, or on one project. - list_users (read-only): The instance's user directory. OpenProject restricts it to administrators, so an ordinary account gets a 403 — that is a permission answer, not a broken connection. For a project's people, list_project_members needs no admin rights. - list_saved_views (read-only): The saved views (queries) people work from — each one's name and the project it belongs to. Useful to answer 'what does the team actually look at'. - list_work_package_attachments (read-only): Files attached to a work package: name, size, content type and who uploaded them. - list_work_package_watchers (read-only): Who is watching a work package — the people OpenProject notifies of every change on it. - add_work_package_watcher (write, needs approval): Add someone as a watcher of a work package, so OpenProject notifies them of every change. Writes to OpenProject, so it goes through the confirmation guardrail when enabled. Example prompts: - "List work packages, optionally within one project. `statusFilter` open/closed is the usual cut; ids for the other filters come from list_statuses, list_types, list_priorities and list_project_members — from OpenProject." - "Get one work package. Returns its `lockVersion` — the number an update needs — alongside its fields, with links resolved to {id, name} rather than raw HAL hrefs — from OpenProject." - "The activity trail of one work package: comments and the field changes that were made, oldest first — from OpenProject." --- ## Pennylane URL: https://www.proxoria.com/en/connectors/pennylane Connect your AI to Pennylane. Pennylane accounting for one company: read customer & supplier invoices, the customer/supplier directories, the general ledger (journals, accounts, entries, trial balance, fiscal years), banking (accounts, transactions, invoice payments) and subscriptions; and reconcile a bank transaction against an invoice. Authentication: connect Pennylane with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_customer_invoices (read-only): List the company's customer (sales) invoices, most recent first. Paginated with a cursor. - get_customer_invoice (read-only): Get the details of a single customer (sales) invoice by its id. - list_supplier_invoices (read-only): List the company's supplier (purchase) invoices, most recent first. Paginated with a cursor. - get_supplier_invoice (read-only): Get the details of a single supplier (purchase) invoice by its id. - list_customers (read-only): List the company's customers (the sales directory). Paginated with a cursor. - list_suppliers (read-only): List the company's suppliers (the purchase directory). Paginated with a cursor. - list_journals (read-only): List the accounting journals (sales, purchases, bank, etc.). - list_ledger_accounts (read-only): List the chart of accounts (ledger accounts) with their numbers and labels. - list_ledger_entries (read-only): List general-ledger entries (accounting entries), most recent first. Paginated with a cursor. - get_ledger_entry (read-only): Get one general-ledger entry by its id. - get_trial_balance (read-only): Get the trial balance (balance générale) for a period: each account with its debit, credit and balance. - list_fiscal_years (read-only): List the company's fiscal years (accounting periods). - list_bank_accounts (read-only): List the company's bank accounts. - list_transactions (read-only): List bank transactions, most recent first. Paginated with a cursor. - get_transaction (read-only): Get one bank transaction by its id, including the invoices matched to it. - list_invoice_payments (read-only): List the payments recorded against a customer or supplier invoice (its settlement history). - list_billing_subscriptions (read-only): List recurring billing subscriptions (abonnements) invoiced to customers. - match_transaction (write, needs approval): Reconcile a bank transaction with an invoice: match the given transaction to a customer or supplier invoice. Writes to Pennylane, so it goes through the confirmation guardrail when enabled. Example prompts: - "List the company's customer (sales) invoices, most recent first. Paginated with a cursor — from Pennylane." - "Get the details of a single customer (sales) invoice by its id — from Pennylane." - "List the company's supplier (purchase) invoices, most recent first. Paginated with a cursor — from Pennylane." --- ## Pennylane (Firm) URL: https://www.proxoria.com/en/connectors/pennylane-firm Connect your AI to Pennylane (Firm). Pennylane Firm API for an accounting firm: list your portfolio of client companies, then read each company's customer/supplier directories, general ledger (entries, trial balance, fiscal years), banking (accounts, transactions) and accounting categories. Read-only. Authentication: connect Pennylane (Firm) with a secure OAuth connection; credentials are encrypted at rest. This connector is read-only; it never modifies data. Tools: - list_companies (read-only): List the client companies in the firm's portfolio, with their id and name. Use an id here as the company_id for the other tools. Paginated with a cursor. - list_customers (read-only): List a client company's customers (the sales directory). Paginated with a cursor. - list_suppliers (read-only): List a client company's suppliers (the purchase directory). Paginated with a cursor. - get_trial_balance (read-only): Get a client company's trial balance (balance générale) for a period: each account with its debit, credit and balance. - list_ledger_entries (read-only): List a client company's general-ledger entries (accounting entries). Paginated with a cursor. - list_fiscal_years (read-only): List a client company's fiscal years (accounting periods). - list_bank_accounts (read-only): List a client company's bank accounts. - list_transactions (read-only): List a client company's bank transactions. Paginated with a cursor. - list_categories (read-only): List a client company's analytical categories. - list_category_groups (read-only): List a client company's analytical category groups. Example prompts: - "List the client companies in the firm's portfolio, with their id and name. Use an id here as the company_id for the other tools. Paginated with a cursor — from Pennylane (Firm)." - "List a client company's customers (the sales directory). Paginated with a cursor — from Pennylane (Firm)." - "List a client company's suppliers (the purchase directory). Paginated with a cursor — from Pennylane (Firm)." --- ## Personio URL: https://www.proxoria.com/en/connectors/personio Connect your AI to Personio. Personio HR on API v2: people and their employments, compensation, absences and attendance, projects and time booked to them, the organisation chart, recruiting and custom reports. Authentication: connect Personio with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_employees (read-only): List the people of the company, with their employments. Filter by email or by name to find someone without knowing their id. - get_employee (read-only): Get one person, with their personal details and current employment. - create_employee (write, needs approval): Create a person and their first employment — an onboarding. Only the name is strictly required, but an employment without a start date, status and type is of little use, so give those together. Needs the persons:write scope on the Personio credential. - update_employee (write, needs approval): Change a person's own details. Their employment is changed with update_employment. - list_employments (read-only): List a person's employments — their contracts over time, with start and end dates, working hours, legal entity, job and manager. - get_employment (read-only): Get one employment of a person in full. - update_employment (write, needs approval): Change a contract: status, dates, working hours, job, manager or legal entity. This is the record of someone's terms of employment — a wrong value here is visible to payroll. - list_compensations (read-only): List compensation entries — salaries and other paid components, each with its amount, currency, interval and the date it takes effect. Personio refuses a date range wider than one month, so ask month by month. - create_compensation (write, needs approval): Record a compensation for a person — a raise, a bonus, a new salary. Takes effect on `effectiveFrom` and feeds payroll. The type comes from list_compensation_types. - list_compensation_types (read-only): List the kinds of compensation the company has defined — base salary, bonus, allowance. Their ids are what create_compensation needs. - list_salary_bands (read-only): List the salary bands defined for the company. - get_salary_band (read-only): Get one salary band, with its range. - list_absences (read-only): List absence periods — holidays, sick leave, and every other kind of time off, each with its type, dates and who is standing in. - get_absence (read-only): Get one absence period in full. - list_absence_types (read-only): List the kinds of absence the company has defined — paid leave, sick leave, unpaid. Their ids are what create_absence needs. - create_absence (write, needs approval): Book an absence for someone. Dates are given without a timezone; use `startHalf` or `endHalf` when the absence covers only half of the first or last day. Whether a stand-in is required, optional or forbidden depends on the absence type's own policy. - update_absence (write, needs approval): Change the dates, stand-in or comment of an absence already booked. - list_attendances (read-only): List attendance periods — the hours people actually worked, and their breaks, optionally booked against a project. - get_attendance (read-only): Get one attendance period in full. - create_attendance (write, needs approval): Record worked time or a break for someone. A period cannot exceed 24 hours, and a break cannot be booked against a project — only work can. - update_attendance (write, needs approval): Change the times, project or comment of an attendance period already recorded. - list_projects (read-only): List the projects time can be booked against. - get_project (read-only): Get one project. - create_project (write, needs approval): Create a project people can book time against. - update_project (write, needs approval): Rename a project, or close it to new time bookings. - list_project_members (read-only): List the people who can book time against a project. - add_project_members (write, needs approval): Let people book time against a project. - remove_project_members (write, needs approval): Stop people booking time against a project. Time already booked stays — only the right to add more is removed. - list_org_units (read-only): List the organisational units — departments and teams. - get_org_unit (read-only): Get one organisational unit, with its place in the hierarchy. - list_legal_entities (read-only): List the legal entities of the group — the companies people are employed by. - get_legal_entity (read-only): Get one legal entity. - list_cost_centers (read-only): List the cost centres employments can be charged to. - list_workplaces (read-only): List the workplaces — the offices and sites people are attached to. - list_jobs (read-only): List the jobs defined for the company — the roles people are hired into. - get_job (read-only): Get one job. - list_documents (read-only): List the metadata of employee documents — contracts, certificates, payslips. Returns what each document is, not its content. - list_applications (read-only): List job applications, with the candidate and the job each one is for. - get_application (read-only): Get one application in full. - list_application_stages (read-only): List an application's stage transitions — when it moved to each step of the hiring process, which is how you measure where candidates stall. - list_candidates (read-only): List the candidates who have applied. - get_candidate (read-only): Get one candidate's details. - list_open_positions (read-only): List the positions open to applications. These are recruiting jobs, distinct from the company's job catalogue returned by list_jobs. - get_open_position (read-only): Get one open position. - list_job_categories (read-only): List the categories open positions are grouped under. - list_reports (read-only): List the custom reports built in Personio. Running one is how you answer a question the other tools do not cover, using the company's own definitions. - get_report (read-only): Get one report with its data. - list_report_attributes (read-only): List the attributes a report can be built from — the company's own fields, including the custom ones. Read this to know what Personio actually holds before asking for it. Example prompts: - "List the people of the company, with their employments. Filter by email or by name to find someone without knowing their id — from Personio." - "Get one person, with their personal details and current employment — from Personio." - "List a person's employments — their contracts over time, with start and end dates, working hours, legal entity, job and manager — from Personio." --- ## Pipedrive URL: https://www.proxoria.com/en/connectors/pipedrive Connect your AI to Pipedrive. Pipedrive CRM (API v2): deals, persons, organizations, products, activities, projects and tasks — read, search, create, update and delete — plus pipelines, stages and the custom-field definitions behind them. Authentication: connect Pipedrive with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_activities (read-only): List activities. - create_activity (write, needs approval): Create an activity. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_activity (write, needs approval): Delete an activity. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_activity (read-only): Get an activity by id. - update_activity (write, needs approval): Update an activity. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_project_boards (read-only): List project boards. - create_project_board (write, needs approval): Create a project board. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_project_board (write, needs approval): Delete a project board. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_project_board (read-only): Get a project board by id. - update_project_board (write, needs approval): Update a project board. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_deals (read-only): List deals. - create_deal (write, needs approval): Create a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_archived_deals (read-only): List archived deals — those set aside without being deleted. - list_installments_across_deals (read-only): List the installments of SEVERAL deals at once, passing their ids. Installments are the payment schedule of a deal. - list_products_across_deals (read-only): List the products attached to SEVERAL deals at once, passing their ids. For one deal, use list_deal_products. - search_deals (read-only): Search deals. - delete_deal (write, needs approval): Delete a deal. Pipedrive keeps it recoverable from the trash for 30 days, then removes it for good. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_deal (read-only): Get a deal by id. - update_deal (write, needs approval): Update a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - convert_deal_to_lead (write, needs approval): Convert a deal into a lead. Conversion is asynchronous: the response carries a conversion id to poll with get_deal_conversion_status. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_deal_conversion_status (read-only): Poll the status of a deal-to-lead conversion started by convert_deal_to_lead. - list_deal_discounts (read-only): List discounts added to a deal. - add_deal_discount (write, needs approval): Add a discount to a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_deal_discount (write, needs approval): Delete a discount from a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - update_deal_discount (write, needs approval): Update a discount added to a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - add_deal_installment (write, needs approval): Add an installment to a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_deal_installment (write, needs approval): Delete an installment from a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - update_deal_installment (write, needs approval): Update an installment added to a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_deal_products_bulk (write, needs approval): Delete several products from a deal in one call. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_deal_products (read-only): List products attached to a deal. - add_deal_product (write, needs approval): Add a product to a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - add_deal_products_bulk (write, needs approval): Add several products to a deal in one call. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_deal_product (write, needs approval): Delete an attached product from a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - update_deal_product (write, needs approval): Update the product attached to a deal. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - search_items (read-only): Search across several item types at once (deals, persons, organizations, products, files…) and get the matches ranked by relevance. - search_items_by_field (read-only): Find the values a single field takes across records — the way to answer 'which values exist for this field?'. - search_leads (read-only): Search leads. - convert_lead_to_deal (write, needs approval): Convert a lead into a deal. Conversion is asynchronous: the response carries a conversion id to poll with get_lead_conversion_status. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_lead_conversion_status (read-only): Poll the status of a lead-to-deal conversion started by convert_lead_to_deal. - list_organizations (read-only): List organizations. - create_organization (write, needs approval): Create an organization. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - search_organizations (read-only): Search organizations. - delete_organization (write, needs approval): Delete an organization. Recoverable from Pipedrive's trash for 30 days, then gone. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_organization (read-only): Get an organization by id. - update_organization (write, needs approval): Update an organization. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_persons (read-only): List persons. - create_person (write, needs approval): Create a person. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - search_persons (read-only): Search persons. - delete_person (write, needs approval): Delete a person. Recoverable from Pipedrive's trash for 30 days, then gone. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_person (read-only): Get a person by id. - update_person (write, needs approval): Update a person. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_person_picture (read-only): Get the profile picture of a person. - list_project_phases (read-only): List the phases of a project board. - create_project_phase (write, needs approval): Create a project phase. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_project_phase (write, needs approval): Delete a project phase. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_project_phase (read-only): Get a project phase by id. - update_project_phase (write, needs approval): Update a project phase. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_pipelines (read-only): List pipelines. - create_pipeline (write, needs approval): Create a pipeline. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_pipeline (write, needs approval): Delete a pipeline. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_pipeline (read-only): Get a pipeline by id. - update_pipeline (write, needs approval): Update a pipeline. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_products (read-only): List products. - create_product (write, needs approval): Create a product. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - search_products (read-only): Search products. - delete_product (write, needs approval): Delete a product. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_product (read-only): Get a product by id. - update_product (write, needs approval): Update a product. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - duplicate_product (write, needs approval): Duplicate a product, copying its fields into a new one. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_product_image (write, needs approval): Delete the image of a product. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_product_image (read-only): Get the image attached to a product. - list_product_variations (read-only): List product variations. - create_product_variation (write, needs approval): Create a product variation. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_product_variation (write, needs approval): Delete a product variation. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - update_product_variation (write, needs approval): Update a product variation. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_project_templates (read-only): List project templates. - get_project_template (read-only): Get a template by id. - list_projects (read-only): List projects. - create_project (write, needs approval): Create a project. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_archived_projects (read-only): List archived projects. - search_projects (read-only): Search projects. - delete_project (write, needs approval): Delete a project. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_project (read-only): Get a project by id. - update_project (write, needs approval): Update a project. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - archive_project (write, needs approval): Archive a project: it leaves the active lists without being deleted. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_project_changelog (read-only): List the changes made to a project's field values, newest first. - list_project_permitted_users (read-only): List the users allowed to see a project. - list_stages (read-only): List stages. - create_stage (write, needs approval): Create a stage. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_stage (write, needs approval): Delete a stage. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_stage (read-only): Get a stage by id. - update_stage (write, needs approval): Update stage details. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_tasks (read-only): List tasks. - create_task (write, needs approval): Create a task. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_task (write, needs approval): Delete a task. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - get_task (read-only): Get a task by id. - update_task (write, needs approval): Update a task. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - list_fields (read-only): List the field definitions of one entity — their keys, labels, types and options. The key is what every custom field is read and written by, so start here whenever a custom field is involved. - get_field (read-only): Get one field definition of an entity by its field code. - create_field (write, needs approval): Create a custom field on an entity. This changes the CRM's schema for everyone, not one record. `options` is required for the enum and set types. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - update_field (write, needs approval): Rename a custom field or change its description. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_field (write, needs approval): Delete a custom field definition. This is a schema change: the field disappears from every record of that entity, with the values it held. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - add_field_options (write, needs approval): Add option labels to an enum or set field, in bulk. Existing options are kept. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - update_field_options (write, needs approval): Rename existing options of an enum or set field, in bulk. The whole call fails if any option id is unknown. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. - delete_field_options (write, needs approval): Delete options of an enum or set field, in bulk. Records holding a deleted option lose that value. The whole call fails if any option id is unknown. Writes to Pipedrive, so it goes through the confirmation guardrail when enabled. Example prompts: - "List activities — from Pipedrive." - "Get an activity by id — from Pipedrive." - "List project boards — from Pipedrive." --- ## PrestaShop URL: https://www.proxoria.com/en/connectors/prestashop Connect your AI to PrestaShop. Your own PrestaShop store: browse orders with their status history, customers and their addresses, products, variants, price rules, stock and categories, spot low stock, read CMS pages and the customer-service threads, and reach the shop's reference tables (carriers, currencies, countries, taxes…); adjust stock quantities and move an order to another status. Connects to the shop's webservice with a webservice key. Authentication: connect PrestaShop with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_orders (read-only): List orders, newest first. Filter by status id (see list_order_states), customer or creation date. Returns a compact view; use get_order for one order's full detail. - get_order (read-only): Get one order in full: rows (products, quantities, prices), totals, carrier, payment and current status. - list_order_states (read-only): List the shop's order statuses (id → label). Orders carry a status id; this is the mapping that makes them readable. - list_customers (read-only): Search customer accounts by email or last name. - get_customer (read-only): Get one customer account in detail. - list_products (read-only): List catalog products, optionally filtered by name or reference. - get_product (read-only): Get one product in full: prices, description, categories, combinations. - list_categories (read-only): List the shop's categories. - get_stock (read-only): Get the stock rows of one product (one per combination). - list_low_stock (read-only): List stock rows whose quantity has fallen to or below a threshold — the restocking view. - update_stock_quantity (write, needs approval): Set the stock quantity of a product (or one of its combinations). Writes to the shop, so it goes through the confirmation guardrail when enabled. - list_cms_pages (read-only): List the shop's CMS pages (legal, about, delivery…), read-only. - list_addresses (read-only): Postal addresses on file, optionally those of one customer — where an order actually ships to. - list_order_history (read-only): The status history of an order: every state it went through and when — what tells whether a late order is stuck at payment or at shipping. - update_order_state (write, needs approval): Move an order to another status (ids from list_order_states). PrestaShop treats this as a real event: it can email the customer and generate an invoice, so it is a write and goes through the confirmation guardrail when enabled. - list_product_combinations (read-only): The variants (combinations) of a product — size, colour… — with their own reference, price impact and quantity. - list_specific_prices (read-only): Price rules in force: discounts by product, customer group, country or date range — why a customer is not paying the catalogue price. - list_customer_threads (read-only): Customer service threads with their status and channel — the shop's inbox for questions about orders. - list_customer_messages (read-only): The messages inside one customer service thread, oldest first. - list_reference_data (read-only): List one of the shop's reference tables, to get the ids and labels the other tools report: carriers, currencies, languages, countries, states, zones, taxes, manufacturers, suppliers, stores or employees. Example prompts: - "List orders, newest first. Filter by status id (see list_order_states), customer or creation date. Returns a compact view; use get_order for one order's full detail — from PrestaShop." - "Get one order in full: rows (products, quantities, prices), totals, carrier, payment and current status — from PrestaShop." - "List the shop's order statuses (id → label). Orders carry a status id; this is the mapping that makes them readable — from PrestaShop." --- ## Redmine URL: https://www.proxoria.com/en/connectors/redmine Connect your AI to Redmine. Your own Redmine: browse and search issues with their history, projects and their members, versions and categories, time entries, wiki pages, news, files and attachments; create and update issues, comment them, relate them to each other, add watchers, log time and write wiki pages. Connects to your instance's REST API with a Redmine API key. Authentication: connect Redmine with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_issues (read-only): List issues, filtered. Ids come from the lookup tools: list_projects, list_issue_statuses, list_trackers, list_project_members. Use `statusFilter` for the common open/closed cut rather than guessing status ids. - get_issue (read-only): Get one issue in full: fields, custom fields, its journal (comments and every field change) and its relations. Custom fields come back as {id, name, value} — that is where to learn a custom field's id before writing it with update_issue. - search (read-only): Full-text search across the instance — issues, wiki pages, projects, documents. Use it when you do not know which project or resource holds the answer; use list_issues when you do. - create_issue (write, needs approval): Create an issue. `projectId`, `trackerId` and `priorityId` come from list_projects, list_trackers and list_enumerations — a tracker must be enabled on the project or Redmine refuses the whole call. Writes to Redmine, so it goes through the confirmation guardrail when enabled. - update_issue (write, needs approval): Update an issue's fields. Only what you pass changes. To add a comment without touching any field, use add_issue_note instead. - add_issue_note (write, needs approval): Add a comment to an issue, changing nothing else. Set `private` for a note only visible to members with the right permission. - list_issue_relations (read-only): Relations of one issue — what it blocks, what blocks it, what it precedes, duplicates or relates to. The way to answer 'what is holding this up?'. - create_issue_relation (write, needs approval): Relate two issues. `precedes` and `follows` accept a delay in days; the other types do not. Writes to Redmine. - list_projects (read-only): List the projects visible to this key, with their id, name and identifier. - get_project (read-only): Get one project with the trackers enabled on it and its issue categories — needed before creating an issue, since a tracker not enabled on the project is refused. - list_project_members (read-only): Who is on a project, with their user id and roles. This is how you turn a person's name into the id that assignedToId expects — Redmine's global user directory is administrator-only, this is not. - list_trackers (read-only): Trackers of the instance (Bug, Feature, Support…), with their ids. A tracker must also be enabled on the target project — get_project says which are. - list_issue_statuses (read-only): Issue statuses of the instance with their ids, and which of them count as closed. - list_enumerations (read-only): Enumerated values with their ids: issue priorities, time-tracking activities or document categories. Needed before creating an issue or logging time. - list_time_entries (read-only): Logged time, filterable by project, issue, user or date range. - log_time (write, needs approval): Log time against an issue or a project. `activityId` comes from list_enumerations (kind time_entry_activities). Writes to Redmine. - list_wiki_pages (read-only): Index of a project's wiki: page titles, versions and parents. - get_wiki_page (read-only): Get one wiki page with its text and current version number — pass that version to update_wiki_page so a concurrent edit is refused rather than overwritten. - create_wiki_page (write, needs approval): Create a wiki page. Fails if the title already exists — use update_wiki_page for an existing one. Writes to Redmine. - update_wiki_page (write, needs approval): Replace a wiki page's text. Pass the `version` returned by get_wiki_page: Redmine then refuses the write if someone edited the page meanwhile, instead of silently discarding their work. Writes to Redmine. - list_versions (read-only): Versions (milestones) of a project, with their due date and status — the ids `fixed_version` points at on an issue. - list_issue_categories (read-only): Issue categories defined on a project, with their ids and assignee. - list_news (read-only): Announcements posted on the instance, or on one project. - list_project_files (read-only): Files attached to a project (releases, specs…), with size and download count. - list_users (read-only): The instance's user directory — administrator-only in Redmine. For a project's people without admin rights, use list_project_members. - get_my_account (read-only): Who this API key belongs to: name, login, email and admin flag. Useful to check what the connection can actually see before blaming an empty result on the data. - get_attachment (read-only): Metadata of one attachment — filename, size, content type, author and its download URL (ids come from get_issue, which includes the issue's attachments). - add_issue_watcher (write, needs approval): Add someone as a watcher of an issue, so Redmine notifies them of every change. Writes to Redmine. - list_reference_data (read-only): Instance-wide configuration the other tools need ids from: roles, user groups, custom field definitions, or the saved queries (filters) people work from. Example prompts: - "List issues, filtered. Ids come from the lookup tools: list_projects, list_issue_statuses, list_trackers, list_project_members. Use `statusFilter` for the common open/closed cut rather than guessing status ids — from Redmine." - "Get one issue in full: fields, custom fields, its journal (comments and every field change) and its relations. Custom fields come back as {id, name, value} — that is where to learn a custom field's id before writing it with update_issue — from Redmine." - "Full-text search across the instance — issues, wiki pages, projects, documents. Use it when you do not know which project or resource holds the answer; use list_issues when you do — from Redmine." --- ## Ringover URL: https://www.proxoria.com/en/connectors/ringover Connect your AI to Ringover. Ringover cloud telephony: browse call logs, contacts and phone numbers, send SMS and read the replies, and reach what was actually said on a call — transcriptions, searchable by meaning, plus Empower's AI summaries and key moments. Authentication: connect Ringover with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_calls (read-only): List recent calls (paginated). Optionally filter by date range. - get_call (read-only): Get a single call by its id, with its details. - list_contacts (read-only): List contacts, optionally filtered by a search term. - list_numbers (read-only): List the phone numbers of the Ringover account. - send_sms (write, needs approval): Send an SMS from one of the account's numbers. Writes to the real world, so it goes through the confirmation guardrail when enabled. - list_transcriptions (read-only): List call transcriptions, or search their content. Given a `search` term, Ringover runs a semantic search over what was actually said and returns the closest matches — the way to answer 'which calls mentioned the renewal?'. In search mode the date and duration filters are ignored. Use get_transcription for the full text of one known call. - get_transcription (read-only): Get the full transcription of one call, as `transcription_data.speeches`. In each speech, `channelId` is 1 for the person who placed the call and 0 for the person who received it. Takes the call id from list_calls, not an Empower UUID. - list_analyzed_calls (read-only): List the calls analysed by Empower, Ringover's AI layer, and get their `call_uuid` — the identifier the other Empower tools need. Filter by call date, direction or source. Ringover refuses a date range wider than 15 days. - get_call_analysis (read-only): Get everything Empower knows about one analysed call: transcription, scores and metadata. Takes the `call_uuid` from list_analyzed_calls. - get_call_summary (read-only): Get Empower's AI summary of a call, with its key topics, action items and sentiment. Read `recap.summary` — the top-level `summary` field is a retired pipeline and is always null. Takes the `call_uuid` from list_analyzed_calls. - get_call_moments (read-only): Get the key moments Empower detected during a call — objections, commitments, questions — each with where it occurred. Takes the `call_uuid` from list_analyzed_calls. - list_conversations (read-only): List SMS and chat conversations. Without monitoring enabled on the API key, only the conversations of the key's own user are returned. - list_conversation_messages (read-only): Read the messages of one conversation — including the replies to an SMS sent with send_sms. Pages with `lastIdReturned` rather than an offset, so no message is seen twice. - list_tags (read-only): List the call tags defined for the team, with their names and colours. Tags are shared by every user of the team. - list_users (read-only): List the team's Ringover users, to attribute a call to the person who took it. Without monitoring enabled on the API key, only the key's own user is returned. - get_contact (read-only): Get one contact: names, company, numbers and whether it is shared with the team. - create_contacts (write, needs approval): Create contacts, up to fifty in one call. Each is filed under the user the API key belongs to. Note that this route does double duty at Ringover: the same POST with a filter body reads contacts instead, which is what list_contacts uses. - update_contact (write, needs approval): Replace a contact's details. This is a full replacement: the numbers you send become the contact's numbers, so send back the ones that should survive. - set_contact_company (write, needs approval): Change only a contact's company, leaving everything else untouched. Ringover carries the new value in the path, so it must be a single path segment. - set_contact_firstname (write, needs approval): Change only a contact's first name, leaving everything else untouched. Ringover carries the new value in the path, so it must be a single path segment. - set_contact_lastname (write, needs approval): Change only a contact's last name, leaving everything else untouched. Ringover carries the new value in the path, so it must be a single path segment. - add_contact_numbers (write, needs approval): Add one or more phone numbers to a contact, keeping the ones already there. - update_contact_number (write, needs approval): Change one of a contact's numbers — the value itself, its type, or both. - get_conversation (read-only): Get one conversation: its name, its kind, and when it was last active. - list_conversation_members (read-only): List who is in a conversation. - rename_conversation (write, needs approval): Rename a conversation. Everyone in it sees the new name. - get_user (read-only): Get one user: their name, numbers, profile and current status. - get_user_planning (read-only): A user's weekly working hours, day by day. What decides whether a call reaches them or goes to the fallback. - list_presences (read-only): Who is available, busy, on a call or away, across the whole team, right now. - get_user_presence (read-only): One user's current availability. - list_profiles (read-only): List the permission profiles of the account — what each class of user is allowed to do. - get_profile (read-only): Get one permission profile and its rights. - get_team (read-only): The account itself: its name, its plan and its overall settings. - get_number (read-only): Get one phone number of the account: what it is called, where it routes, and whether it is active. - list_ivrs (read-only): List the interactive voice menus of the account. An IVR is the tree a caller hears before reaching anyone, and it is where a routing problem usually lives. - get_ivr (read-only): Get one voice menu with its numbers and its configuration. - list_ivr_scenarios (read-only): List the scenarios of one voice menu. A scenario is one routing plan — office hours, holidays, out of hours — and only one is active at a time. - get_ivr_scenario (read-only): Get one scenario of a voice menu, with the steps a caller goes through. - list_scenarios (read-only): List every scenario of the account, across all voice menus. - get_scenario (read-only): Get one scenario by id, whichever menu it belongs to. - list_conferences (read-only): List the conference rooms of the account, with their numbers and access codes. - get_conference (read-only): Get one conference room. - list_groups (read-only): List the call groups — the rings that distribute an incoming call across several people. - get_group (read-only): Get one call group with its members and how the call is spread among them. - list_team_tasks (read-only): List the team's tasks — the follow-ups attached to calls. Ringover reads these through a POST because the filter is too big for a URL; nothing is created or changed by this call. - create_tasks (write, needs approval): Create a batch of tasks from a list of numbers — one task per number, all sharing the same type, status and scheduled date. - list_blacklisted_numbers (read-only): List the numbers blocked for the whole team, with the comment left on each. - add_blacklisted_number (write, needs approval): Block a number for the whole team. Nobody in the account will take or place a call to it in the direction you choose. - get_blacklisted_number (read-only): Check whether a number is on the team blacklist, and read the comment if it is. - set_blacklist_comment (write, needs approval): Change the comment on a team-blocked number. Ringover carries the text in the path, so it must be a single path segment. - list_user_blacklisted_numbers (read-only): List the numbers one user has blocked for themselves, which is separate from the team list. - add_user_blacklisted_number (write, needs approval): Block a number for one user only, leaving the rest of the team unaffected. - get_user_blacklisted_number (read-only): Check whether a number is on one user's personal blacklist. - set_user_blacklist_comment (write, needs approval): Change the comment on a number one user has blocked. - start_callback (write, needs approval): Place a call. Ringover rings the chosen device first, then dials the destination and joins the two — so this really does make somebody's phone ring, twice over, and cannot be recalled. - start_ivr_callback (write, needs approval): Place a call through a voice menu, so the destination sees the menu's number and the call is logged against it. Same warning as start_callback: a phone actually rings. - list_call_campaigns (read-only): List the outbound call campaigns of the account. A campaign is an automatic dialler: it walks a list of numbers and connects each answer to an available agent. - list_my_call_campaigns (read-only): List the active campaigns assigned to the user the API key belongs to. - get_call_campaign (read-only): Get one campaign: its state, its schedule, its retry policy and how far through its list it is. - get_campaign_logs (read-only): The audit trail of a campaign — who started it, stopped it, changed it, and when. - list_campaign_numbers (read-only): List the contacts in a campaign's queue, with their position, their attempt count and the custom columns carried alongside them. - add_campaign_numbers (write, needs approval): Add contacts to a campaign's queue. Once the campaign is running, these numbers will be dialled. - update_campaign_numbers (write, needs approval): Change the custom column values carried by contacts already in the queue. - remove_campaign_numbers (write, needs approval): Take contacts out of a campaign's queue so they are not called. - list_campaign_users (read-only): List the agents assigned to a campaign — the people its calls are handed to. - assign_campaign_users (write, needs approval): Assign agents to a campaign. - remove_campaign_users (write, needs approval): Take agents off a campaign. - list_campaign_groups (read-only): List the call groups assigned to a campaign. - assign_campaign_groups (write, needs approval): Assign call groups to a campaign, which assigns everyone in them. - remove_campaign_groups (write, needs approval): Take call groups off a campaign. - add_campaign_column (write, needs approval): Add a custom data column to a campaign's contacts — a field the agent sees on screen when the call connects. - remove_campaign_column (write, needs approval): Delete a custom data column and its values from every contact of the campaign. - rename_campaign_column (write, needs approval): Rename a custom data column, keeping its values. - switch_campaign_columns (write, needs approval): Swap two custom columns, which changes the order the agent sees them in. - set_campaign_max_retry (write, needs approval): Set how many times the dialler retries a number that did not answer. - rename_call_campaign (write, needs approval): Rename a campaign. - start_call_campaign (write, needs approval): Start a campaign. From that moment the dialler works through the queue and rings real people, one number after another, until it is stopped or the list runs out. Give `startDate` to have it begin later instead of now. - edit_campaign_start (write, needs approval): Move the scheduled start of a campaign that has not begun yet. - stop_call_campaign (write, needs approval): Stop a running campaign. Calls in progress finish; nothing new is dialled. - reset_call_campaign (write, needs approval): Reset a campaign to its initial state: every contact goes back into the queue as if never attempted, and the attempt counters are cleared. The call history itself is not affected. - set_campaign_script (write, needs approval): Attach a call script to a campaign, or detach the current one by leaving `scriptUuid` out. - set_campaign_tags (write, needs approval): Replace a campaign's tags. THIS REPLACES THE WHOLE SET: the tags you send become the campaign's tags, so send back the ones that should stay. - remove_campaign_tags (write, needs approval): Remove every tag from a campaign. - remove_campaign_tag (write, needs approval): Remove one tag from a campaign, leaving the others. Example prompts: - "List recent calls (paginated). Optionally filter by date range — from Ringover." - "Get a single call by its id, with its details — from Ringover." - "List contacts, optionally filtered by a search term — from Ringover." --- ## Sellsy URL: https://www.proxoria.com/en/connectors/sellsy Connect your AI to Sellsy. Sellsy CRM and invoicing: search the account, browse companies, contacts, opportunities, estimates, invoices and payments, and create companies, contacts, opportunities, tasks and comments. Authentication: connect Sellsy with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - search (read-only): Full-text search across the whole Sellsy account (companies, contacts, opportunities, invoices…). Optionally restrict to object types. - list_companies (read-only): List companies (clients and prospects), paginated. - get_company (read-only): Get a company by id (details, addresses, smart tags). - create_company (write, needs approval): Create a company (prospect or client). Writes to Sellsy, so it goes through the confirmation guardrail when enabled. - list_contacts (read-only): List contacts (people attached to companies), paginated. - create_contact (write, needs approval): Create a contact (a person). Writes to Sellsy, so it goes through the confirmation guardrail when enabled. - list_opportunities (read-only): List opportunities (deals), paginated. - get_opportunity (read-only): Get an opportunity by id (amount, status, pipeline step, related party). - list_pipelines (read-only): List the opportunity pipelines. Their ids (and their steps' ids) are required by create_opportunity. - create_opportunity (write, needs approval): Create an opportunity linked to a company or individual, in a pipeline step (get ids from list_pipelines). Writes to Sellsy, so it goes through the confirmation guardrail when enabled. - update_opportunity (write, needs approval): Update an opportunity's simple fields: name, status (open/won/lost/cancelled/late) or pipeline step. Writes to Sellsy, so it goes through the confirmation guardrail when enabled. - list_estimates (read-only): List estimates (quotes/devis), paginated. Read-only. - get_estimate (read-only): Get an estimate by id (amounts, status, linked party). Read-only. - list_invoices (read-only): List invoices, paginated. Read-only. - get_invoice (read-only): Get an invoice by id (amounts, payment status, due date, linked party). Read-only. - list_payments (read-only): List received payments, paginated. Read-only. - create_task (write, needs approval): Create a task, optionally linked to a company, individual or opportunity. Writes to Sellsy, so it goes through the confirmation guardrail when enabled. - add_comment (write, needs approval): Add a comment to a Sellsy object's timeline (company, individual or opportunity). Writes to Sellsy, so it goes through the confirmation guardrail when enabled. Example prompts: - "Full-text search across the whole Sellsy account (companies, contacts, opportunities, invoices…). Optionally restrict to object types — from Sellsy." - "List companies (clients and prospects), paginated — from Sellsy." - "Get a company by id (details, addresses, smart tags) — from Sellsy." --- ## Shopify URL: https://www.proxoria.com/en/connectors/shopify Connect your AI to Shopify. Your Shopify store: orders and their edits, draft orders, fulfilment, returns and refunds, products, collections and inventory, customers and segments, B2B companies, discounts and gift cards, blog, pages and navigation, translations, metaobjects, Shopify Payments payouts and disputes, and ShopifyQL analytics. Every write waits for your approval. Authentication: connect Shopify with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_abandoned_checkouts (read-only): Returns a list of abandoned checkouts. A checkout is considered abandoned when a customer adds contact information but doesn't complete their purchase. Includes both abandoned and recovered checkouts. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_abandoned_checkouts (read-only): Returns the count of abandoned checkouts for the given shop. Limited to a maximum of 10000 by default. - get_order (read-only): The `order` query retrieves an order by its ID. This query provides access to comprehensive order information such as customer details, line items, financial data, and fulfillment status. - get_order_by_identifier (read-only): Return an order by an identifier. - get_order_payment_status (read-only): Retrieves the status of a deferred payment by its payment reference ID. Use this query to monitor the processing status of payments that are initiated through payment mutations. Deferred payments are called payment terms in the API. - list_orders (read-only): Returns a list of orders placed in the store, including data such as order status, customer, and line item details. Use the `orders` query to build reports, analyze sales performance, or automate fulfillment workflows. The `orders` query supports pagination, sorting, and filtering. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_orders (read-only): Returns the number of orders in the shop. You can filter orders using search syntax or a `SavedSearch`, and set a maximum count limit to control query performance. - count_pending_orders (read-only): The number of pendings orders. Limited to a maximum of 10000. - get_refund (read-only): Retrieves a refund by its ID. A refund represents a financial record of money returned to a customer from an order. It provides a comprehensive view of all refunded amounts, transactions, and restocking instructions associated with returning products or correcting order issues. - list_tender_transactions (read-only): Transactions representing a movement of money between customers and the shop. Each transaction records the amount, payment method, processing details, and the associated `Order`. Returns one page: pass pageInfo.endCursor as `after` for the next. - cancel_order (write, needs approval): Cancels an order, with options for refunding, restocking inventory, and customer notification. Depending on refundMethod and restock, cancelling also refunds and restocks; notifyCustomer emails the customer. It cannot be undone. - capture_payment (write, needs approval): Captures payment for an authorized transaction on an order. Use this mutation to claim the money that was previously reserved by an authorization transaction. Charges the customer's card for an amount that was only authorized. The charge is real. - close_order (write, needs approval): Marks an open `Order` as closed. A closed order is one where merchants fulfill or cancel all `LineItem` objects and complete all financial transactions. - remove_order_customer (write, needs approval): Removes customer from an order. - set_order_customer (write, needs approval): Sets a customer on an order. - send_order_invoice (write, needs approval): Sends an email invoice for an `Order`. Emails the customer. - mark_order_as_paid (write, needs approval): Marks an order as paid by recording a payment transaction for the outstanding amount. Records a payment received outside Shopify (cash, bank transfer). No money moves. - open_order (write, needs approval): Opens a closed order. - update_order (write, needs approval): Updates the attributes of an order, such as the customer's email, the shipping address for the order, tags, and metafields associated with the order. - send_payment_reminder (write, needs approval): Sends an email payment reminder for a payment schedule. Emails the customer. - refund_order (write, needs approval): Creates a refund for an order, allowing you to process returns and issue payments back to customers. Money goes back to the customer's payment method and cannot be taken back; pass notify to email them. - edit_order (write, needs approval): Edit an existing order in one go: add variants or custom items, change quantities, add, update or remove discounts and shipping lines. The changes are staged in an edit session and committed together; if any change is refused, nothing is committed and the order is left as it was. Lines added in this call cannot be referenced by a later change of the same call. notifyCustomer emails the customer the updated order. - get_draft_order (read-only): Retrieves a draft order by its ID. A draft order is an order created by a merchant on behalf of their customers. Draft orders contain all necessary order details (products, pricing, customer information) but require payment to be accepted before they can be converted into completed orders. - list_draft_order_delivery_options (read-only): Available delivery options for a `DraftOrder` based on the provided input. The query returns shipping rates, local delivery rates, and pickup locations that merchants can choose from when creating draft orders. - get_draft_order_tag (read-only): Returns a `DraftOrderTag` resource by ID. - list_draft_orders (read-only): List of saved draft orders. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_draft_orders (read-only): Returns the number of draft orders that match the query. Limited to a maximum of 10000 by default. - add_draft_order_tags (write, needs approval): Adds tags to multiple draft orders. - remove_draft_order_tags (write, needs approval): Removes tags from multiple draft orders. - calculate_draft_order (read-only): Calculates the properties of a `DraftOrder` without creating it. Returns pricing information including `CalculatedDraftOrderLineItem` totals, shipping charges, applicable discounts, and tax calculations based on the provided `Customer` and `MailingAddress` information. - complete_draft_order (write, needs approval): Completes a draft order and converts it into a regular order. The order appears in the merchant's orders list, and the customer can be notified about their order. Turns the draft into a real order. - create_draft_order (write, needs approval): Creates a draft order with attributes such as customer information, line items, shipping and billing addresses, and payment terms. Draft orders are useful for merchants that need to: - create_draft_order_from_order (write, needs approval): Creates a draft order from order. - delete_draft_order (write, needs approval): Deletes a draft order. - duplicate_draft_order (write, needs approval): Duplicates a draft order. - preview_draft_order_invoice (read-only): Previews a draft order invoice email. - send_draft_order_invoice (write, needs approval): Sends an invoice email for a `DraftOrder`. The invoice includes a secure checkout link for reviewing and paying for the order. Use the `email` argument to customize the email, such as the subject and message. Emails the customer a checkout link. - update_draft_order (write, needs approval): Updates a draft order. - get_fulfillment (read-only): Retrieves a `Fulfillment` by its ID. A fulfillment is a record that the merchant has completed their work required for one or more line items in an `Order`. It includes tracking information, `LineItem` objects, and the status of the fulfillment. - get_fulfillment_order (read-only): Returns a `FulfillmentOrder` resource by ID. - list_fulfillment_orders (read-only): The paginated list of all fulfillment orders. The returned fulfillment orders are filtered according to the fulfillment order access scopes granted to the app. Returns one page: pass pageInfo.endCursor as `after` for the next. - list_held_fulfillment_orders (read-only): Returns a list of fulfillment orders that are on hold. Returns one page: pass pageInfo.endCursor as `after` for the next. - cancel_fulfillment (write, needs approval): Cancels an existing `Fulfillment` and reverses its effects on associated `FulfillmentOrder` objects. When you cancel a fulfillment, the system creates new fulfillment orders for the cancelled items so they can be fulfilled again. - create_fulfillment (write, needs approval): Creates a fulfillment for one or more `FulfillmentOrder` objects. The fulfillment orders are associated with the same `Order` and are assigned to the same `Location`. - create_fulfillment_event (write, needs approval): Creates a `FulfillmentEvent` to track the shipment status and location of items that have shipped. Events capture status updates like carrier pickup, in transit, out for delivery, or delivered. Needs the write_fulfillments scope, which Shopify grants to fulfilment services only. Without it the call is refused; update_fulfillment_tracking covers ordinary tracking updates. - cancel_fulfillment_order (write, needs approval): Cancels a `FulfillmentOrder` and creates a replacement fulfillment order to represent the work left to be done. The original fulfillment order will be marked as closed. - hold_fulfillment_order (write, needs approval): Applies a fulfillment hold on a fulfillment order. - mark_ready_for_pickup (write, needs approval): Marks fulfillment order line items as ready for customer pickup. When executed, this mutation automatically sends a "Ready For Pickup" notification to the customer. - merge_fulfillment_order (write, needs approval): Merges a set or multiple sets of fulfillment orders together into one based on line item inputs and quantities. - move_fulfillment_order (write, needs approval): Changes the location which is assigned to fulfill a number of unfulfilled fulfillment order line items. - release_fulfillment_order_hold (write, needs approval): Releases the fulfillment hold on a fulfillment order. - reschedule_fulfillment_order (write, needs approval): Reschedules a scheduled fulfillment order. - split_fulfillment_order (write, needs approval): Splits `FulfillmentOrder` objects by moving the specified `LineItem` objects and quantities into a new fulfillment order. - set_fulfillment_deadline (write, needs approval): Sets the latest date and time by which the fulfillment orders need to be fulfilled. - update_fulfillment_tracking (write, needs approval): Updates tracking information for a fulfillment, including the carrier name, tracking numbers, and tracking URLs. You can provide either single or multiple tracking numbers for shipments with multiple packages. - get_return (read-only): Retrieves a return by its ID. A return represents the intent of a buyer to ship one or more items from an order back to a merchant or a third-party fulfillment location. - calculate_return (read-only): Calculates the financial outcome of a `Return` without creating it. Use this query to preview return costs before initiating the actual return process. - list_return_reasons (read-only): Returns the full library of available return reason definitions. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_returnable_fulfillment (read-only): Returns a `ReturnableFulfillment` resource by ID. - list_returnable_fulfillments (read-only): List of returnable fulfillments. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_reverse_delivery (read-only): Lookup a reverse delivery by ID. - get_reverse_fulfillment_order (read-only): Lookup a reverse fulfillment order by ID. - remove_from_return (write, needs approval): Removes return and/or exchange lines from a return. - approve_return_request (write, needs approval): Approves a customer's return request. If this mutation is successful, then the `Return.status` field of the approved return is set to `OPEN`. - cancel_return (write, needs approval): Cancels a return and restores the items back to being fulfilled. Canceling a return is only available before any work has been done on the return (such as an inspection or refund). - close_return (write, needs approval): Indicates a return is complete, either when a refund has been made and items restocked, or simply when it has been marked as returned in the system. - create_return (write, needs approval): Creates a return from an existing order that has at least one fulfilled line item that hasn't yet been refunded. If you create a return on an archived order, then the order is automatically unarchived. - decline_return_request (write, needs approval): Declines a return on an order. When a return is declined, each `ReturnLineItem.fulfillmentLineItem` can be associated to a new return. Use the `ReturnCreate` or `ReturnRequest` mutation to initiate a new return. - process_return (write, needs approval): Processes a return by confirming which items customers return and exchange, handling their disposition, and optionally issuing refunds. This mutation confirms the quantities for `ReturnLineItem` and `ExchangeLineItem` objects previously created on the `Return`. Can issue the refund in the same step. - reopen_return (write, needs approval): Reopens a closed return. - request_return (write, needs approval): Creates a return request that requires merchant approval before processing. The return has its status set to `REQUESTED` and the merchant must approve or decline it. - create_reverse_delivery (write, needs approval): Creates a new reverse delivery with associated external shipping information. - update_reverse_delivery_shipping (write, needs approval): Updates a reverse delivery with associated external shipping information. - dispose_returned_items (write, needs approval): Disposes reverse fulfillment order line items. - get_product (read-only): Retrieves a product by its ID. A product is an item that a merchant can sell in their store. - get_product_by_identifier (read-only): Return a product by an identifier. - get_product_duplicate_job (read-only): Returns the product duplicate job. - get_product_operation (read-only): Returns a ProductOperation resource by ID. - list_product_tags (read-only): Returns tags added to `Product` objects in the shop. Provides a paginated list of tag strings. Returns one page: pass pageInfo.endCursor as `after` for the next. - list_product_types (read-only): Returns a paginated list of product types assigned to products in the store. The maximum page size is 1000. The maximum page size is 1000. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_product_variant (read-only): Retrieves a product variant by its ID. - get_product_variant_by_identifier (read-only): Return a product variant by an identifier. - list_product_variants (read-only): Retrieves a list of product variants associated with a product. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_product_variants (read-only): Count of product variants. Limited to a maximum of 10000 by default. - list_product_vendors (read-only): The list of vendors added to products. The maximum page size is 1000. Returns one page: pass pageInfo.endCursor as `after` for the next. - list_products (read-only): Retrieves a list of products in a store. Products are the items that merchants can sell in their store. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_products (read-only): Count of products. Limited to a maximum of 10000 by default. - get_selling_plan_group (read-only): Returns a `SellingPlanGroup` resource by ID. - list_selling_plan_groups (read-only): Retrieves a paginated list of `SellingPlanGroup` objects that belong to the app making the API call. Selling plan groups are selling methods like subscriptions, preorders, or other purchase options that merchants offer to customers. Returns one page: pass pageInfo.endCursor as `after` for the next. - search_product_taxonomy (read-only): Access to Shopify's standardized product taxonomy for categorizing products. The `Taxonomy` organizes products into a hierarchical tree structure with categories, attributes, and values. - create_product_bundle (write, needs approval): Creates a product bundle that groups multiple `Product` objects together as components. The bundle appears as a single product in the store, with its price determined by the parent product and inventory calculated from the component products. - update_product_bundle (write, needs approval): Updates a product bundle or componentized product. - create_product (write, needs approval): Creates a product with attributes such as title, description, vendor, and media. - delete_product (write, needs approval): Permanently deletes a product and all its associated data, including variants, media, publications, and inventory items. Permanent, with its variants and media. - duplicate_product (write, needs approval): Duplicates a product. - update_product_option (write, needs approval): Updates an option on a product, such as size, color, or material. Each option includes a name, position, and a list of values. The combination of a product option and value creates a product variant. - create_product_options (write, needs approval): Creates one or more options on a product, such as size, color, or material. Each option includes a name, position, and a list of values. The combination of a product option and value creates a product variant. - delete_product_options (write, needs approval): Deletes one or more options from a product. Product options define the choices available for a product, such as size, color, or material. - reorder_product_options (write, needs approval): Reorders the options and option values on a product, updating the order in which product variants are presented to customers. - reorder_product_media (write, needs approval): Reorders media attached to a product, changing their sequence in product displays. The operation processes asynchronously to handle products with large media collections. - update_product (write, needs approval): Updates a product with attributes such as title, description, vendor, and media. - attach_variant_media (write, needs approval): Appends existing media from a product to specific variants of that product, creating associations between media files and particular product options. This allows different variants to showcase relevant images or videos. - detach_variant_media (write, needs approval): Detaches media from product variants. - create_product_variants (write, needs approval): Creates multiple product variants for a single product in one operation. You can run this mutation directly or as part of a bulk operation for large-scale catalog updates. - delete_product_variants (write, needs approval): Deletes multiple variants in a single `Product`. Specify the product ID and an array of variant IDs to remove variants in bulk. You can call this mutation directly or through the `bulkOperationRunMutation` mutation. Returns the updated product and any `UserError` objects. - reorder_product_variants (write, needs approval): Reorders multiple variants in a single product. This mutation can be called directly or via the bulkOperation. - update_product_variants (write, needs approval): Updates multiple product variants for a single product in one operation. You can run this mutation directly or as part of a bulk operation for large-scale catalog updates. - add_tags (write, needs approval): Adds tags to a resource. If the resource type doesn't support tagging, the `id` argument returns a resource-not-found error. - remove_tags (write, needs approval): Removes tags from a resource. If the resource type doesn't support tagging, the `id` argument returns a resource-not-found error. - get_collection (read-only): Retrieves a collection by its ID. A collection represents a grouping of products that merchants can display and sell as a group in their online store and other sales channels. - get_collection_by_identifier (read-only): Return a collection by an identifier. - list_collections (read-only): Retrieves a list of collections in a store. Collections are groups of products that merchants can organize for display in their online store and other sales channels. For example, an athletics store might create different collections for running attire, shoes, and accessories. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_collections (read-only): Count of collections. Limited to a maximum of 10000 by default. - create_collection (write, needs approval): Creates a collection to group products together in the online store and other sales channels. For example, an athletics store might create different collections for running attire, shoes, and accessories. - delete_collection (write, needs approval): Deletes a collection and removes it permanently from the store. This operation cannot be undone and will remove the collection from all sales channels where it was published. Permanent. The products stay. - duplicate_collection (write, needs approval): Duplicates a collection. - reorder_collection_products (write, needs approval): Asynchronously reorders products within a specified collection. Instead of returning an updated collection, this mutation returns a job, which should be polled. The `Collection.sortOrder` must be `MANUAL`. - update_collection (write, needs approval): Updates a collection, modifying its properties, products, or publication settings. Collections help organize products together in the online store and other sales channels. - get_publication (read-only): Retrieves a `Publication` by `ID`. - list_publications (read-only): Returns a paginated list of `Publication`. Returns one page: pass pageInfo.endCursor as `after` for the next. - publish_resource (write, needs approval): Publishes a resource, such as a `Product` or `Collection`, to one or more publications. - unpublish_resource (write, needs approval): Unpublishes a resource, such as a `Product` or `Collection`, from one or more publications. The resource remains in your store but becomes unavailable to customers. - get_inventory_item (read-only): Returns an InventoryItem object by ID. - list_inventory_items (read-only): Returns a list of inventory items. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_inventory_level (read-only): Returns an InventoryLevel object by ID. - get_inventory_shipment (read-only): Retrieves an `InventoryShipment` by ID. Returns tracking details, `InventoryShipmentLineItem` objects with quantities, and the shipment's current `InventoryShipmentStatus`. - list_inventory_shipments (read-only): Returns a paginated list of `InventoryShipment` objects. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_inventory_transfer (read-only): Returns an `InventoryTransfer` by ID. Inventory transfers track the movement of inventory between locations, including origin and destination details, `InventoryTransferLineItem` objects, quantities, and `InventoryTransferStatus` values. - list_inventory_transfers (read-only): Returns a paginated list of `InventoryTransfer` objects between locations. Transfers track the movement of `InventoryItem` objects between `Location` objects. Returns one page: pass pageInfo.endCursor as `after` for the next. - activate_inventory_at_location (write, needs approval): Activates an inventory item at a `Location` by creating an `InventoryLevel` that tracks stock quantities. This enables you to manage inventory for a `ProductVariant` at the specified location. - adjust_inventory (write, needs approval): Adjusts quantities for inventory items by applying incremental changes at specific locations. Each adjustment modifies the quantity by a delta value rather than setting an absolute amount. Shopify compares before it writes: read the current quantity with get_inventory_item or get_inventory_level and pass it as changeFromQuantity, or the change is refused. - deactivate_inventory_at_location (write, needs approval): Removes an inventory item's quantities from a location, and turns off inventory at the location. Refused when it would leave the item stocked nowhere: a product must remain stocked at one location at least. - update_inventory_item (write, needs approval): Updates an `InventoryItem`'s properties including whether inventory is tracked, cost, SKU, and whether shipping is required. Inventory items represent the goods available to be shipped to customers. - move_inventory (write, needs approval): Moves inventory quantities for a single inventory item between different states at a single location. Use this mutation to reallocate inventory across quantity states without moving it between locations. Shopify compares before it writes: pass changeFromQuantity on each side, read from get_inventory_item or get_inventory_level. - set_inventory (write, needs approval): Set quantities of specified name using absolute values. This mutation supports compare-and-set functionality to handle concurrent requests properly. If `ignoreCompareQuantity` is not set to true, the mutation will only update the quantity if the persisted quantity matches the `compareQuantity` value. If the `compareQuantity` value does not match the persisted value, the mutation will return an error. Absolute quantities. Pass compareQuantity (the quantity you read) so a concurrent change cannot be overwritten silently. - create_inventory_shipment (write, needs approval): Adds a draft shipment to an inventory transfer. - mark_inventory_shipment_in_transit (write, needs approval): Marks a draft inventory shipment as in transit. - receive_inventory_shipment (write, needs approval): Receive an inventory shipment. - set_inventory_shipment_tracking (write, needs approval): Edits the tracking info on an inventory shipment. - cancel_inventory_transfer (write, needs approval): Cancels an inventory transfer. - create_inventory_transfer (write, needs approval): Creates a draft inventory transfer to move inventory items between `Location` objects in your store. The transfer tracks which items to move, their quantities, and the origin and destination locations. - mark_inventory_transfer_ready (write, needs approval): Sets an inventory transfer to ready to ship. - set_inventory_transfer_items (write, needs approval): This mutation sets the quantity for one or more line items on a Transfer. - get_delivery_profile (read-only): Retrieves a `DeliveryProfile` by ID. Delivery profiles group shipping settings for specific `Product` objects that ship from selected `Location` objects to delivery zones. - list_delivery_profiles (read-only): Returns a paginated list of `DeliveryProfile` objects for the shop. Delivery profiles group `Product` and `ProductVariant` objects that share shipping rates and zones. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_location (read-only): Retrieves a `Location` by its ID. Locations are physical places where merchants store inventory, such as warehouses, retail stores, or fulfillment centers. - get_location_by_identifier (read-only): Return a location by an identifier. - list_locations (read-only): A paginated list of inventory locations where merchants can stock `Product` items and fulfill `Order` items. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_locations (read-only): Returns the count of locations for the given shop. Limited to a maximum of 10000 by default. - get_customer (read-only): Returns a `Customer` resource by ID. - get_customer_by_identifier (read-only): Return a customer by an identifier. - get_customer_merge_status (read-only): Returns the status of a customer merge request job. - preview_customer_merge (read-only): Returns a preview of a customer merge request. - list_segment_members (read-only): A paginated list of customers that belong to an individual `Segment`. Segments group customers based on criteria defined through ShopifyQL queries. Access segment members with their profile information and purchase summary data. The connection includes statistics for analyzing segment attributes (such as average and sum calculations) and a total count of all members. The maximum page size is 1000. Returns one page: pass pageInfo.endCursor as `after` for the next. - list_customers (read-only): Returns a list of customers in your Shopify store, including key information such as name, email, location, and purchase history. Use this query to segment your audience, personalize marketing campaigns, or analyze customer behavior by applying filters based on location, order history, marketing preferences and tags. The `customers` query supports pagination and sorting. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_customers (read-only): The number of customers. Limited to a maximum of 10000 by default. - get_segment (read-only): Retrieves a customer `Segment` by ID. Segments are dynamic groups of customers that meet specific criteria defined through ShopifyQL queries. - list_segment_filters (read-only): A list of filters. Returns one page: pass pageInfo.endCursor as `after` for the next. - list_segments (read-only): Returns a paginated list of `Segment` objects for the shop. Segments are dynamic groups of customers that meet specific criteria defined through ShopifyQL queries. You can filter segments by search query and sort them by creation date or other criteria. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_segments (read-only): The number of segments for a shop. Limited to a maximum of 10000 by default. - get_store_credit_account (read-only): Retrieves a `StoreCreditAccount` by ID. Store credit accounts hold monetary balances that account owners can use at checkout. The owner is either a `Customer` or a `CompanyLocation`. - create_customer_address (write, needs approval): Creates a new `MailingAddress` for a `Customer`. You can optionally set the address as the customer's default address. - delete_customer_address (write, needs approval): Deletes a customer's address. - update_customer_address (write, needs approval): Updates a `Customer`'s `MailingAddress`. You can modify any field of the address and optionally set it as the customer's default address. - create_customer (write, needs approval): Creates a new `Customer` in the store. - delete_customer (write, needs approval): Deletes a `Customer` from the store. You can only delete customers who haven't placed any orders. Permanent. Shopify refuses it for a customer who has orders. - update_customer_email_consent (write, needs approval): Updates a `Customer`'s email marketing consent information. The customer must have an email address to update their consent. Records the marketing state (such as subscribed, pending, unsubscribed), opt-in level, and when and where the customer gave or withdrew consent. Record only consent the customer actually gave: it decides who receives marketing email. - merge_customers (write, needs approval): Merges two customers. - send_customer_account_invite (write, needs approval): Sends an email invitation for a customer to create a legacy customer account. The invitation lets customers set up their password and activate their account in the online store. Emails the customer. - update_customer_sms_consent (write, needs approval): Updates a customer's SMS marketing consent information. The customer must have a phone number on their account to receive SMS marketing. Record only consent the customer actually gave: it decides who receives marketing SMS. - update_customer (write, needs approval): Updates a `Customer`'s attributes including personal information and `tax exemptions`. - set_customer_default_address (write, needs approval): Updates a customer's default address. - create_segment (write, needs approval): Creates a segment. - delete_segment (write, needs approval): Deletes a segment. - update_segment (write, needs approval): Updates a segment. - credit_store_credit (write, needs approval): Adds funds to a `StoreCreditAccount` by creating a `StoreCreditAccountCreditTransaction`. The mutation accepts either a store credit account ID, a `Customer` ID, or a `CompanyLocation` ID. When you provide a customer or company location ID, it automatically creates an account if one doesn't exist for the specified currency. Adds store credit the customer can spend. - debit_store_credit (write, needs approval): Creates a debit transaction that decreases the store credit account balance by the given amount. Removes store credit from the customer. - list_companies (read-only): A paginated list of companies in the shop. `Company` objects are business entities that purchase from the merchant. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_companies (read-only): The number of companies for a shop. Limited to a maximum of 10000 by default. - get_company (read-only): Returns a `Company` resource by ID. - get_company_contact (read-only): Returns a `CompanyContact` resource by ID. - get_company_location (read-only): Returns a `CompanyLocation` resource by ID. - list_company_locations (read-only): A paginated list of `CompanyLocation` objects for B2B customers. Company locations represent individual branches or offices of a `Company` where B2B orders can be placed. Returns one page: pass pageInfo.endCursor as `after` for the next. - add_customer_to_company (write, needs approval): Adds an existing `Customer` as a contact to a `Company`. Companies are business entities that make purchases from the merchant's store. Use this mutation when you have a customer who needs to be associated with a B2B company to make purchases on behalf of that company. - assign_company_contact_role (write, needs approval): Assigns a role to a contact for a location. - create_company_contact (write, needs approval): Creates a company contact and the associated customer. - update_company_contact (write, needs approval): Updates a company contact. - create_company (write, needs approval): Creates a `Company` for B2B commerce. This mutation creates the company and can optionally create an initial `CompanyContact` and `CompanyLocation` in a single operation. Company contacts are people who place orders on behalf of the company. Company locations are branches or offices with their own billing and shipping addresses. - create_company_location (write, needs approval): Creates a new location for a `Company`. Company locations are branches or offices where B2B customers can place orders with specific pricing, catalogs, and payment terms. - update_company_location (write, needs approval): Updates a company location's information and B2B checkout settings. Company locations are branches or offices where `CompanyContact` members place orders on behalf of the company. Contacts must be assigned to a location through `roleAssignments` to place orders. - update_company (write, needs approval): Updates a `Company` with new information. Companies represent business customers that can have multiple contacts and locations with specific pricing, payment terms, and checkout settings. - get_discount_by_code (read-only): Retrieves a code discount by its discount code. The search is case-insensitive, enabling you to find discounts regardless of how customers enter the code. - count_discount_codes (read-only): The total number of discount codes for the shop. Limited to a maximum of 10000 by default. - get_discount (read-only): Returns a `DiscountNode` resource by ID. - list_discounts (read-only): Returns a list of discounts. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_discounts (read-only): The total number of discounts for the shop. Limited to a maximum of 10000 by default. - activate_automatic_discount (write, needs approval): Activates an automatic discount. - create_automatic_discount (write, needs approval): Creates an amount off discount that's automatically applied on a cart and at checkout. - update_automatic_discount (write, needs approval): Updates an existing amount off discount that's automatically applied on a cart and at checkout. - create_bxgy_automatic_discount (write, needs approval): Creates a buy X get Y discount (BXGY) that's automatically applied on a cart and at checkout. - update_bxgy_automatic_discount (write, needs approval): Updates an existing buy X get Y discount (BXGY) that's automatically applied on a cart and at checkout. - deactivate_automatic_discount (write, needs approval): Deactivates an automatic discount. - delete_automatic_discount (write, needs approval): Deletes an existing automatic discount from the store, permanently removing it from all future order calculations. This mutation provides a clean way to remove promotional campaigns that are no longer needed. - create_free_shipping_automatic_discount (write, needs approval): Creates automatic free shipping discounts that apply to qualifying orders without requiring discount codes. These promotions automatically activate when customers meet specified criteria, streamlining the checkout experience. - update_free_shipping_automatic_discount (write, needs approval): Updates existing automatic free shipping discounts, allowing merchants to modify promotion criteria, shipping destinations, and eligibility requirements without recreating the entire discount structure. - activate_code_discount (write, needs approval): Activates a previously created code discount, making it available for customers to use during checkout. This mutation transitions inactive discount codes into an active state where they can be applied to orders. - create_code_discount (write, needs approval): Creates an amount off discount that's applied on a cart and at checkout when a customer enters a code. Amount off discounts can be a percentage off or a fixed amount off. - update_code_discount (write, needs approval): Updates an amount off discount that's applied on a cart and at checkout when a customer enters a code. Amount off discounts can be a percentage off or a fixed amount off. - bulk_activate_discount_codes (write, needs approval): Activates multiple code discounts asynchronously using one of the following: - A search query - A saved search ID - A list of discount code IDs - bulk_deactivate_discount_codes (write, needs approval): Deactivates multiple code-based discounts asynchronously using one of the following: - A search query - A saved search ID - A list of discount code IDs - create_bxgy_code_discount (write, needs approval): Creates a buy X get Y discount (BXGY) that's applied on a cart and at checkout when a customer enters a code. - update_bxgy_code_discount (write, needs approval): Updates a buy X get Y discount (BXGY) that's applied on a cart and at checkout when a customer enters a code. - deactivate_code_discount (write, needs approval): Temporarily suspends a code discount without permanently removing it from the store. Deactivation allows merchants to pause promotional campaigns while preserving the discount configuration for potential future use. - delete_code_discount (write, needs approval): Removes a code discount from the store, making it permanently unavailable for customer use. This mutation provides a clean way to eliminate discount codes that are no longer needed or have been replaced. - create_free_shipping_code_discount (write, needs approval): Creates an free shipping discount that's applied on a cart and at checkout when a customer enters a code. - update_free_shipping_code_discount (write, needs approval): Updates a free shipping discount that's applied on a cart and at checkout when a customer enters a code. - delete_discount_codes (write, needs approval): Asynchronously delete discount codes in bulk that customers can use to redeem a discount. - add_discount_codes (write, needs approval): Asynchronously add discount codes in bulk that customers can use to redeem a discount. You can use the `discountRedeemCodeBulkAdd` mutation to automate the distribution of discount codes through emails or other marketing channels. - get_gift_card (read-only): Retrieves a `GiftCard` by its ID. Returns the gift card's balance, transaction history, `Customer` information, and whether it's enabled. - list_gift_cards (read-only): Returns a paginated list of `GiftCard` objects issued for the shop. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_gift_cards (read-only): Returns the total count of gift cards that have been issued by the shop. Use this for dashboard summaries or to understand the scale of a merchant's gift card program. The count includes all gift cards regardless of status (active, disabled, or fully redeemed). Limited to a maximum of 10000 by default. - create_gift_card (write, needs approval): Creates a new `GiftCard` with a specified initial value. You can assign the gift card to a `Customer` or create it without assignment for manual distribution. - credit_gift_card (write, needs approval): Adds funds to an existing gift card, increasing its available balance. Use this when a merchant wants to top up a customer's gift card — for example, as a promotional bonus, a customer service gesture, or to reload a reusable gift card. Adds spendable value to the gift card. - deactivate_gift_card (write, needs approval): Deactivate a gift card. A deactivated gift card cannot be used by a customer. A deactivated gift card cannot be re-enabled. - debit_gift_card (write, needs approval): Removes funds from a gift card, decreasing its available balance. Use this for manual balance adjustments — for example, correcting an accidental over-credit or applying a fee. Removes spendable value from the gift card. - send_gift_card_to_customer (write, needs approval): Sends a notification to the customer who purchased a gift card, including the gift card details and code. The notification is delivered using the customer's available contact method. Use this to resend the purchase confirmation or remind the purchaser about a gift card they bought. Emails the customer. - send_gift_card_to_recipient (write, needs approval): Sends a notification to the designated recipient of a gift card, delivering the gift card code and redemption instructions. The notification is delivered using the recipient's available contact method. Use this to deliver or re-deliver the gift card to the intended recipient. Emails the gift card's recipient. - update_gift_card (write, needs approval): Updates the properties of an existing gift card, such as its expiration date, note, or template suffix. Use this to modify gift card details — for example, extending an expiration date for a loyal customer or adding an internal note for tracking purposes. - list_marketing_activities (read-only): A list of marketing activities associated with the marketing app. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_marketing_activity (read-only): Returns a `MarketingActivity` resource by ID. - get_marketing_event (read-only): Returns a `MarketingEvent` resource by ID. - list_marketing_events (read-only): A list of marketing events associated with the marketing app. Returns one page: pass pageInfo.endCursor as `after` for the next. - run_shopifyql_query (read-only): Executes a ShopifyQL query to analyze store data and returns results in a tabular format. ShopifyQL, e.g. FROM sales SHOW total_sales BY month SINCE -12m. Tabular result. Needs the app's level 2 access to protected customer data. - get_article (read-only): Returns a `Article` resource by ID. - list_article_authors (read-only): List of article authors for the shop. Returns one page: pass pageInfo.endCursor as `after` for the next. - list_article_tags (read-only): List of all article tags. - list_articles (read-only): Returns a paginated list of articles from the shop's blogs. `Article` objects are blog posts that contain content like text, images, and tags. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_blog (read-only): Returns a `Blog` resource by ID. - list_blogs (read-only): Returns a paginated list of the shop's `Blog` objects. Blogs serve as containers for `Article` objects and provide content management capabilities for the store's editorial content. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_blogs (read-only): Count of blogs. Limited to a maximum of 10000 by default. - get_comment (read-only): Returns a `Comment` resource by ID. - list_comments (read-only): List of the shop's comments. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_menu (read-only): Returns a `Menu` resource by ID. - list_menus (read-only): Retrieves navigation menus. Menus organize content into hierarchical navigation structures that merchants can display in the online store (for example, in headers, footers, and sidebars) and customer accounts. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_page (read-only): Returns a `Page` resource by ID. - list_pages (read-only): A paginated list of pages from the online store. `Page` objects are content pages that merchants create to provide information to customers, such as "About Us", "Contact", or policy pages. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_pages (read-only): Count of pages. Limited to a maximum of 10000 by default. - get_url_redirect (read-only): Returns a `UrlRedirect` resource by ID. - list_url_redirects (read-only): A list of redirects for a shop. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_url_redirects (read-only): Count of redirects. Limited to a maximum of 10000 by default. - create_article (write, needs approval): Creates an `Article`. Articles are content pieces that include a title, body text, and author information. - delete_article (write, needs approval): Permanently deletes a blog article from a shop's blog. This mutation removes the article and all associated metadata. - update_article (write, needs approval): Updates an existing `Article`. You can modify the article's content, metadata, publication status, and associated properties like author information and tags. - create_blog (write, needs approval): Creates a new blog within a shop, establishing a container for organizing articles. - delete_blog (write, needs approval): Permanently deletes a blog from a shop. This mutation removes the blog container and its organizational structure. - update_blog (write, needs approval): Updates an existing blog's configuration and settings. This mutation allows merchants to modify blog properties to keep their content strategy current. - approve_comment (write, needs approval): Approves a pending comment, making it visible to store visitors on the associated blog article. - delete_comment (write, needs approval): Permanently removes a comment from a blog article. - mark_comment_not_spam (write, needs approval): Reverses a spam classification on a comment, restoring it to normal moderation status. This mutation allows merchants to change their decision when a comment has been manually marked as spam. - mark_comment_spam (write, needs approval): Marks a comment as spam, removing it from public view. This mutation enables merchants to quickly handle unwanted promotional content, malicious links, or other spam that appears in blog discussions. - create_menu (write, needs approval): Creates a navigation `Menu` for the online store. Menus organize links that help customers navigate to collections, products, pages, blogs, and custom URLs. - delete_menu (write, needs approval): Deletes a menu. - update_menu (write, needs approval): Updates a `Menu` for display on the storefront. Modifies the menu's title and navigation structure, including nested `MenuItem` objects. You can update the handle for non-default menus. - create_page (write, needs approval): Creates a `Page` for the online store. - delete_page (write, needs approval): Permanently deletes a page from the online store. - update_page (write, needs approval): Updates an existing page's content and settings. - create_url_redirect (write, needs approval): Creates a `UrlRedirect` object. - delete_url_redirect (write, needs approval): Deletes a `UrlRedirect` object. - update_url_redirect (write, needs approval): Updates a URL redirect. - list_files (read-only): Retrieves a paginated list of files that have been uploaded to a Shopify store. Files represent digital assets that merchants can upload to their store for various purposes including product images, marketing materials, documents, and brand assets. Returns one page: pass pageInfo.endCursor as `after` for the next. - create_file (write, needs approval): Creates file assets for a store from external URLs or files that were previously uploaded using the `stagedUploadsCreate` mutation. - delete_file (write, needs approval): Deletes file assets that were previously uploaded to your store. - update_file (write, needs approval): Updates properties, content, and metadata associated with an existing file asset that has already been uploaded to Shopify. - get_theme (read-only): Returns an `OnlineStoreTheme` by its ID. Use this query to retrieve theme metadata and access the theme's `files`, which include templates, assets, translations, and configuration files. - list_themes (read-only): Returns a paginated list of `OnlineStoreTheme` objects for the online store. Themes control the appearance and layout of the storefront. Returns one page: pass pageInfo.endCursor as `after` for the next. - list_shop_locales (read-only): Returns the locales enabled on a shop. Each locale represents a language for translations and determines how content displays to customers in different markets. - get_translatable_resource (read-only): Retrieves a resource that has translatable fields. Returns the resource's `Translation` objects for different locales and markets, along with the original `TranslatableContent` and digest values needed to register new translations. - list_translatable_resources (read-only): Returns a paginated list of `TranslatableResource` objects for a specific resource type. Each resource provides translatable content and digest values needed for the `translationsRegister` mutation. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_translatable_resources_by_ids (read-only): Returns a paginated list of `TranslatableResource` objects for the specified resource IDs. Each resource provides translatable content and digest values needed for the `translationsRegister` mutation. Returns one page: pass pageInfo.endCursor as `after` for the next. - register_translations (write, needs approval): Creates or updates translations for a resource's translatable content. - remove_translations (write, needs approval): Deletes translations. - get_metafield_definition (read-only): Retrieves a `MetafieldDefinition` by its identifier. You can identify a definition using either its owner type, namespace, and key, or its global ID. - list_metafield_definitions (read-only): Returns a list of metafield definitions. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_metaobject (read-only): Retrieves a single `Metaobject` by its global ID. Metaobjects store custom structured data based on defined schemas. The returned metaobject includes its fields with values, display name, handle, and associated metadata like update timestamps and capabilities. - get_metaobject_by_handle (read-only): Retrieves a `Metaobject` by its handle and type. Handles are unique identifiers within a metaobject type. - get_metaobject_definition (read-only): Retrieves a `MetaobjectDefinition` by its global ID. Metaobject definitions provide the structure and fields for metaobjects. - get_metaobject_definition_by_type (read-only): Retrieves a `MetaobjectDefinition` by its type. The type serves as a unique identifier that distinguishes one metaobject definition from another. - list_metaobject_definitions (read-only): Returns a paginated list of all `MetaobjectDefinition` objects configured for the store. Metaobject definitions provide the schema for creating custom data structures composed of individual fields. Each definition specifies the field types, access permissions, and capabilities for `Metaobject` entries of that type. Use this query to discover available metaobject types before creating or querying metaobject entries. Returns one page: pass pageInfo.endCursor as `after` for the next. - list_metaobjects (read-only): Returns a paginated list of `Metaobject` entries for a specific type. Metaobjects are custom data structures that extend Shopify's data model with merchant or app-specific data types. Returns one page: pass pageInfo.endCursor as `after` for the next. - delete_metafields (write, needs approval): Deletes `Metafield` objects in bulk by specifying combinations of owner ID, namespace, and key. - set_metafields (write, needs approval): Sets metafield values. Metafield values will be set regardless if they were previously created or not. - create_metaobject (write, needs approval): Creates a metaobject entry based on an existing `MetaobjectDefinition`. The type must match a definition that already exists in the shop. - delete_metaobject (write, needs approval): Deletes the specified metaobject and its associated metafields. - update_metaobject (write, needs approval): Updates a `Metaobject` with new field values, handle, or capabilities. Metaobjects are custom data structures that extend Shopify's data model. - upsert_metaobject (write, needs approval): Creates or updates a `Metaobject` based on its handle. If a metaobject with the specified handle exists, the mutation updates it with the provided field values. If no matching metaobject exists, the mutation creates a new one. - get_market (read-only): Returns a `Market` resource by ID. - list_markets (read-only): Returns a paginated list of `Market` objects configured for the shop. Markets match buyers based on defined conditions to deliver customized shopping experiences. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_price_list (read-only): Returns a `PriceList` by ID. You can use price lists to specify either fixed prices or adjusted relative prices that override initial `Product` prices. - list_price_lists (read-only): All price lists for a shop. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_dispute (read-only): Returns a `ShopifyPaymentsDispute` resource by ID. - list_disputes (read-only): Returns a paginated list of all Shopify Payments disputes for the shop. Disputes occur when a buyer files a complaint with their payments provider, and the merchant must provide evidence to contest it. Each dispute includes the status, amount, reason, and associated order. Use this to monitor and manage open chargebacks and track dispute resolution outcomes. Returns one page: pass pageInfo.endCursor as `after` for the next. - get_shopify_payments_account (read-only): Returns the Shopify Payments account information for the shop. Includes current balances across all currencies, payout schedules, and bank account configurations. - get_event (read-only): Retrieves a single event by ID. Events chronicle activities in your store such as resource creation, updates, or staff comments. The query returns an `Event` interface of type `BasicEvent` or `CommentEvent`. - list_events (read-only): A paginated list of events that chronicle activities in the store. `Event` is an interface implemented by types such as `BasicEvent` and `CommentEvent` that track actions such as creating `Article` objects, fulfilling `Order` objects, adding `Product` objects, or staff comments on timelines. Returns one page: pass pageInfo.endCursor as `after` for the next. - count_events (read-only): Count of events. Limited to a maximum of 10000. - get_job (read-only): Returns a Job resource by ID. Used to check the status of internal jobs and any applicable changes. - get_shop (read-only): Returns the Shop resource corresponding to the access token used in the request. The Shop resource contains business and store management settings for the shop. Example prompts: - "Returns a list of abandoned checkouts. A checkout is considered abandoned when a customer adds contact information but doesn't complete their purchase. Includes both abandoned and recovered checkouts. Returns one page: pass pageInfo.endCursor as `after` for the next — from Shopify." - "Returns the count of abandoned checkouts for the given shop. Limited to a maximum of 10000 by default — from Shopify." - "The `order` query retrieves an order by its ID. This query provides access to comprehensive order information such as customer details, line items, financial data, and fulfillment status — from Shopify." --- ## SurveyMonkey URL: https://www.proxoria.com/en/connectors/surveymonkey Connect your AI to SurveyMonkey. SurveyMonkey: surveys and their structure, aggregated answer counts and how they move over time, the collectors that gathered them and how each performed, contacts and translations. Nothing here sends an invitation. Authentication: connect SurveyMonkey with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_surveys (read-only): List surveys (paginated). - get_survey_details (read-only): Get a survey's full structure (pages and questions). - list_responses (read-only): List responses to a survey (bulk, paginated). - get_response (read-only): Get a single response with its answers. - get_survey_rollups (read-only): Get how many people chose each possible answer, across the whole survey. This is the tool that answers 'what did the respondents say' — reading responses one by one is slower, costlier, and capped at 25 per survey on the basic plans. - get_page_rollups (read-only): Get the answer counts for one page of a survey. - get_question_rollups (read-only): Get the answer counts for one question — the fastest way to settle a single question without pulling the rest of the survey. - get_survey_trends (read-only): Get how the answers moved over time across the whole survey — the same counts as the rollups, but broken down by period, which is how you see a score improving or slipping. - get_page_trends (read-only): Get the answers of one page over time. - get_question_trends (read-only): Get one question's answers over time. - get_survey (read-only): Get a survey's own record — title, dates, response count, language and folder. Use get_survey_details for its pages and questions. - search_surveys (read-only): Find surveys whose title contains a word, rather than paging the whole list. - create_survey (write, needs approval): Create a survey, empty or from a template. It is created unpublished: no one can answer it until a collector is attached, which is done from SurveyMonkey. - share_survey (write, needs approval): Share a survey with colleagues, by email address, granting them read or edit rights. This grants access to a person, so name only people the customer asked for. - list_survey_shares (read-only): List who a survey has been shared with, and with what rights. - list_survey_folders (read-only): List the folders surveys are filed in. - list_survey_templates (read-only): List the survey templates available — SurveyMonkey's own and the team's. Their ids feed create_survey. - list_survey_categories (read-only): List the categories templates are grouped under. - list_available_languages (read-only): List the languages a survey can be written in. Distinct from list_survey_translations, which returns the languages one particular survey has been translated into. - list_pages (read-only): List a survey's pages. - get_page (read-only): Get one page of a survey, with its title and position. - list_questions (read-only): List the questions on a page, with their type and possible answers. The ids here are what the rollup and trend tools need. - get_question (read-only): Get one question in full — its wording, type, and the exact set of answers it offers, which is what makes a count of answers readable. - list_question_bank (read-only): List SurveyMonkey's library of pre-written questions — wording already tested, worth borrowing rather than inventing. - list_collector_responses (read-only): List the responses that came through one collector — the way to compare how the same survey performed by email against by web link. - list_collector_responses_bulk (read-only): List one collector's responses with all their answers, in bulk. - list_collectors (read-only): List a survey's collectors — each way it was sent out: an email invitation, a web link, an embedded form. Their statistics say which one actually worked. - get_collector (read-only): Get one collector, with its type, status and settings. - get_collector_stats (read-only): Get a collector's figures — sent, opened, started, completed. The response rate of one channel, which the survey's own totals hide. - list_collector_messages (read-only): List the messages of an email collector — the invitation and its reminders, each with its own status and schedule. - get_collector_message (read-only): Get one message of an email collector, with its subject and body. - list_message_recipients (read-only): List who a particular message was addressed to, and what became of each send. - add_message_recipients (write, needs approval): Add recipients to a message that has not been sent yet. Adding them does not send anything — this connector exposes no way to send an invitation, because it leaves for everyone on the list at once and cannot be recalled. A human sends it from SurveyMonkey. - list_collector_recipients (read-only): List everyone a collector has ever written to. - list_contacts (read-only): List the contacts of the account, across every list. - get_contact (read-only): Get one contact, with their fields and current state. - list_contacts_bulk (read-only): List contacts with all their fields at once, rather than one call each. - list_contact_lists (read-only): List the contact lists a survey can be sent to. - get_contact_list (read-only): Get one contact list, with how many contacts it holds. - list_contact_list_contacts (read-only): List the contacts of one list. - list_contact_fields (read-only): List the custom fields a contact can carry, with their names. Read this before making sense of a contact's extra columns. - copy_contact_list (write, needs approval): Copy a contact list, leaving the original untouched. - merge_contact_lists (write, needs approval): Merge another list into this one. The contacts of the source end up in the destination; check with get_contact_list which is which before calling, because the two lists do not come back apart afterwards. - list_survey_translations (read-only): List the languages one survey has been translated into, and how complete each translation is. - get_survey_translation (read-only): Get one translation of a survey — the translated wording, question by question. Example prompts: - "List surveys (paginated) — from SurveyMonkey." - "Get a survey's full structure (pages and questions) — from SurveyMonkey." - "List responses to a survey (bulk, paginated) — from SurveyMonkey." --- ## Teamleader URL: https://www.proxoria.com/en/connectors/teamleader Connect your AI to Teamleader. Teamleader Focus: browse and create contacts and companies, manage deals through their pipeline, read quotations and invoices, add notes, and create tasks. Authentication: connect Teamleader with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_contacts (read-only): List contacts, optionally narrowed by a search term (name, email, phone). - get_contact (read-only): Get one contact's full details by id. - create_contact (write, needs approval): Create a contact. Only the last name is required. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - update_contact (write, needs approval): Update a contact's details. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - list_companies (read-only): List companies, optionally narrowed by a search term. - get_company (read-only): Get one company's full details by id. - create_company (write, needs approval): Create a company. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - update_company (write, needs approval): Update a company's details. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - list_notes (read-only): List the notes attached to one entity (a contact, company, deal, quotation, invoice…). - create_note (write, needs approval): Attach a note to an entity (contact, company, deal…). Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - list_deals (read-only): List deals, optionally narrowed by a search term. - get_deal (read-only): Get one deal by id (phase, value, customer, responsible user). - create_deal (write, needs approval): Create a deal for a contact or a company. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - move_deal (write, needs approval): Move a deal to another pipeline phase. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - close_deal (write, needs approval): Close a deal as won or lost. A lost deal requires a reason id. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - list_quotations (read-only): List quotations, optionally those of a single deal. - get_quotation (read-only): Get one quotation by id (status, total, line items). - send_quotation (write, needs approval): Send a quotation to its customer. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - list_invoices (read-only): List invoices, optionally only the drafts or only the booked ones. - get_invoice (read-only): Get one invoice by id (customer, totals, payment status). - book_invoice (write, needs approval): Book a draft invoice, which finalises it and assigns its number. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. - list_tasks (read-only): List tasks, optionally only the open or the completed ones. - create_task (write, needs approval): Create a task, optionally linked to a contact, company or deal. Writes to Teamleader, so it goes through the confirmation guardrail when enabled. Example prompts: - "List contacts, optionally narrowed by a search term (name, email, phone) — from Teamleader." - "Get one contact's full details by id — from Teamleader." - "List companies, optionally narrowed by a search term — from Teamleader." --- ## WordPress URL: https://www.proxoria.com/en/connectors/wordpress Connect your AI to WordPress. Your own WordPress site: search it, browse and write posts and pages (as drafts), moderate and answer comments, manage categories and tags, read the media library, users, revisions, plugins and site settings, and reach the custom content types your theme and plugins add. Connects to the site's REST API with an application password. Authentication: connect WordPress with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - search (read-only): Full-text search across the site's content (posts, pages…). Returns matching items with their type and id. - list_posts (read-only): List blog posts (published by default), optionally filtered by full-text search or status (draft, pending, publish). - get_post (read-only): Get a post by id (title, content, status, categories). - create_post (write, needs approval): Create a blog post as a DRAFT — a human reviews and publishes from WordPress. Writes to the site, so it goes through the confirmation guardrail when enabled. - update_post (write, needs approval): Update a post's title, content or excerpt. Deliberately cannot change the status: publishing stays a human action in WordPress. Writes to the site, so it goes through the confirmation guardrail when enabled. - list_pages (read-only): List the site's pages, optionally filtered by full-text search. - get_page (read-only): Get a page by id (title, content, parent). - create_page (write, needs approval): Create a page as a DRAFT — a human reviews and publishes from WordPress. Pages are structural site content (home, about, legal…). Writes to the site, so it goes through the confirmation guardrail when enabled. - update_page (write, needs approval): Update a page's title or content. Deliberately cannot change the status: publishing stays a human action in WordPress. Writes to the site, so it goes through the confirmation guardrail when enabled. - list_comments (read-only): List comments, optionally those of one post. - list_categories (read-only): List the site's categories (ids are used by create_post). - list_media (read-only): List the media library (images, files), optionally filtered by search. - get_media (read-only): Get one media item: its file URL, sizes, alt text and the post it is attached to. - list_tags (read-only): List the site's tags (ids are used by create_post and update_post). - create_term (write, needs approval): Create a category or a tag. Writes to the site, so it goes through the confirmation guardrail when enabled. - list_users (read-only): List the site's users — the authors a post can be attributed to, with their roles when the credential is allowed to see them. - create_comment (write, needs approval): Post a comment on an article. It follows the site's moderation rules, so it may land as pending rather than published. Writes to the site, so it goes through the confirmation guardrail when enabled. - moderate_comment (write, needs approval): Moderate a comment: approve it, put it back on hold, mark it as spam or send it to the trash. Writes to the site, so it goes through the confirmation guardrail when enabled. - list_post_revisions (read-only): List the saved revisions of a post — who changed it and when — so an edit can be traced or compared. - trash_post (write, needs approval): Send a post to the trash. WordPress keeps it there rather than deleting it, so it can be restored from the admin. Writes to the site, so it goes through the confirmation guardrail when enabled. - get_site_settings (read-only): Read the site's own settings: title, tagline, language, timezone, date format, default category and whether new posts are open to comments. - list_plugins (read-only): List the plugins installed on the site with their version and whether they are active (needs an administrator credential; WordPress refuses it otherwise). - list_content_types (read-only): List the content types this site registers — posts and pages, plus whatever its theme and plugins add (products, events, portfolios…). Each one's `rest_base` is what list_custom_content takes. - list_custom_content (read-only): List entries of a custom content type, for the long tail a dedicated tool would not cover: pass the `rest_base` that list_content_types reported (e.g. 'product', 'tribe_events'). Reads only. Example prompts: - "Full-text search across the site's content (posts, pages…). Returns matching items with their type and id — from WordPress." - "List blog posts (published by default), optionally filtered by full-text search or status (draft, pending, publish) — from WordPress." - "Get a post by id (title, content, status, categories) — from WordPress." --- ## Yousign URL: https://www.proxoria.com/en/connectors/yousign Connect your AI to Yousign. Yousign electronic signatures: browse signature requests and their signers, track status, and activate a draft request to send it out. Authentication: connect Yousign with your API key; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - list_signature_requests (read-only): List signature requests (paginated), optionally filtered by status. - get_signature_request (read-only): Get a signature request by id (documents, signers, status). - list_signers (read-only): List the signers of a signature request (their status and order). - activate_signature_request (write, needs approval): Activate a draft signature request — this sends it to the signers and starts the signing flow. Writes to Yousign, so it goes through the confirmation guardrail when enabled. Example prompts: - "List signature requests (paginated), optionally filtered by status — from Yousign." - "Get a signature request by id (documents, signers, status) — from Yousign." - "List the signers of a signature request (their status and order) — from Yousign." --- ## Zeliq URL: https://www.proxoria.com/en/connectors/zeliq Connect your AI to Zeliq. Zeliq B2B enrichment: find a prospect's professional email or mobile phone from their name and company (or LinkedIn URL), via Zeliq's 40+ source waterfall, and check your credit balance. Authentication: connect Zeliq with your API key; credentials are encrypted at rest. This connector is read-only; it never modifies data. Tools: - check_credit_balance (read-only): Get the remaining Zeliq enrichment credits on the connected account. - enrich_email (read-only): Find a person's professional email from their first name, last name and company (name or domain), or from their LinkedIn URL. Consumes Zeliq credits. Runs in the background: returns a jobId — call get_enrichment_result with it in a few seconds to get the email. Provide at least first_name + last_name + company, or linkedin_url. - enrich_phone (read-only): Find a person's mobile phone number from their LinkedIn URL or their work email. Consumes Zeliq credits. Runs in the background: returns a jobId — call get_enrichment_result with it in a few seconds to get the phone. Provide linkedin_url or email. - get_enrichment_result (read-only): Fetch the result of an enrich_email or enrich_phone job by its jobId. If it reports the job is still running, retry in a few seconds. A result is available for about 15 minutes and returned only once. Example prompts: - "Get the remaining Zeliq enrichment credits on the connected account — from Zeliq." - "Find a person's professional email from their first name, last name and company (name or domain), or from their LinkedIn URL. Consumes Zeliq credits. Runs in the background: returns a jobId — call get_enrichment_result with it in a few seconds to get the email. Provide at least first_name + last_name + company, or linkedin_url — from Zeliq." - "Find a person's mobile phone number from their LinkedIn URL or their work email. Consumes Zeliq credits. Runs in the background: returns a jobId — call get_enrichment_result with it in a few seconds to get the phone. Provide linkedin_url or email — from Zeliq." --- ## Zendesk URL: https://www.proxoria.com/en/connectors/zendesk Connect your AI to Zendesk. Your Zendesk: search and read tickets with their conversation, requesters, organizations and the reference data (groups, views, macros, fields); create and update tickets, and comment them — as an internal note unless a public reply is explicitly asked for. Also searches and drafts help-centre articles. Nothing is deleted, merged or published without a human. Authentication: connect Zendesk with a secure OAuth connection; credentials are encrypted at rest. This connector can make changes; write actions go through a confirmation/approval guardrail before anything is written. Tools: - search_tickets (read-only): Search tickets with Zendesk's own search syntax, e.g. `status:open assignee:none`, `requester:a@b.com created>2026-01-01`, or plain words. This is the tool to reach for first — it answers most questions in one call. - list_tickets (read-only): List tickets, most recently updated first. Prefer search_tickets when filtering. - get_ticket (read-only): Read one ticket. Use list_ticket_comments for the conversation itself. - list_ticket_comments (read-only): The ticket's conversation, oldest first. `public: true` is what the requester saw; `public: false` is an internal note only agents can read. - create_ticket (write, needs approval): Open a ticket. The requester is an existing user id, or a name+email pair for someone not yet in Zendesk. Ids for group and assignee come from list_groups and search_users. - update_ticket (write, needs approval): Change a ticket's status, priority, assignment or tags. To reply or leave a note, use add_ticket_comment instead. - add_ticket_comment (write, needs approval): Add a comment to a ticket. It is an INTERNAL NOTE by default, visible only to agents. Set public to true only when the intent is to reply to the requester: Zendesk emails a public comment to them immediately, and it cannot be taken back. - search_users (read-only): Find agents and end users by name, email, or Zendesk search syntax such as `role:agent` or `organization:Acme`. Ids from here feed assigneeId and requesterId. - get_user (read-only): Read one user (agent or end user) by id. - list_organizations (read-only): List organizations (the companies requesters belong to). - get_organization (read-only): Read one organization by id. - list_groups (read-only): Agent groups — the ids to route a ticket to. - list_ticket_fields (read-only): Ticket fields, including this account's custom ones and their allowed values — what a ticket can actually carry here. - list_views (read-only): Saved views — how this team slices its queues. - list_macros (read-only): Active macros — the team's approved canned responses. Read-only here: use one as the basis for a reply rather than inventing wording. - list_brands (read-only): Brands configured on the account (multi-brand help centres and addresses). - search_articles (read-only): Full-text search across the help centre. - get_article (read-only): Read one help-centre article, with its body. - list_sections (read-only): Help-centre sections — where an article can be filed. - create_article (write, needs approval): Write a help-centre article. It is created as a DRAFT: a person publishes it from Zendesk. sectionId comes from list_sections. - update_article (write, needs approval): Update an existing help-centre article's title or body. Its published state is left exactly as it was — this never publishes and never unpublishes. - list_satisfaction_ratings (read-only): Satisfaction ratings left by requesters, most recent first. Example prompts: - "Search tickets with Zendesk's own search syntax, e.g. `status:open assignee:none`, `requester:a@b.com created>2026-01-01`, or plain words. This is the tool to reach for first — it answers most questions in one call — from Zendesk." - "List tickets, most recently updated first. Prefer search_tickets when filtering — from Zendesk." - "Read one ticket. Use list_ticket_comments for the conversation itself — from Zendesk." --- ## Zoom Webinars URL: https://www.proxoria.com/en/connectors/zoom-webinar Connect your AI to Zoom Webinars. Manage Zoom Webinars: list webinars, view registrants and participants. Authentication: connect Zoom Webinars with a secure OAuth connection; credentials are encrypted at rest. This connector is read-only; it never modifies data. Tools: - list_webinars (read-only): List the authenticated user's Zoom Webinars. `type` selects scheduled or upcoming webinars. - get_webinar (read-only): Get the details of a Zoom Webinar. - list_registrants (read-only): List the registrants of a Zoom Webinar. - list_participants (read-only): List the participants (attendees) who actually joined a past Zoom Webinar, with their join/leave times. Pass the webinar UUID of the past session. Reads the webinar directly (no account-admin or reporting privilege required). Example prompts: - "List the authenticated user's Zoom Webinars. `type` selects scheduled or upcoming webinars — from Zoom Webinars." - "Get the details of a Zoom Webinar — from Zoom Webinars." - "List the registrants of a Zoom Webinar — from Zoom Webinars." --- ## Coming soon (users can vote to prioritize) - Factorial: https://www.proxoria.com/en/connectors/factorial - WooCommerce: https://www.proxoria.com/en/connectors/woocommerce