# Mintlify — MCP server on mcp.ai
> Connect your Mintlify account and use 14 tools for developer tools straight from your AI agent. Connect with your own API key. Mintlify is a documentation platform. This toolkit exposes the Admin API for managing docs programmatically, trigger deployments and previews, run automations, drive the docs-editing agent, and export analytics (feedback, searches, views, visitors and assistant conversations).

By: mcp.ai · official
Page: https://mcp.ai/mintlify

## Connect (MCP protocol)
Remote MCP endpoint (HTTP, streamable): https://api.mcp.ai/p_mintlify?ms=1787296020000
Add it as a custom/remote MCP connector, then authenticate when prompted.

## REST API (no MCP client required)
Every tool is also a REST endpoint, authed with a workspace API key.
Discover: GET https://api.mcp.ai/api/mintlify/_endpoints   # public; lists every endpoint
Call:     POST https://api.mcp.ai/api/mintlify/<endpoint>
  Authorization: Bearer sk_live_…   # create one at https://mcp.ai/settings/api-keys
  Content-Type: application/json
  Body: { …args }   →   { "ok": true, "tool": "<id>", "result": { … } }

## Developer docs
How to use (MCP or REST), markdown: https://mcp.ai/mintlify/skill.md
Postman collection (v2.1): https://mcp.ai/mintlify/postman.json

## Tools
- mintlify_create_agent_job(prompt: string, project_id: string) — Start a background documentation agent job from a natural-language prompt. The agent edits the connected docs repo and OPENS A PULL REQUEST if it changes files (DESTRUCTIVE; consumes credits). Returns
- mintlify_get_agent_job(job_id: string, project_id: string) — Retrieve the status and details (status, PR link, model, source repo/ref) of a docs agent job. Poll this until status is completed, archived or failed. No pagination.
- mintlify_get_assistant_caller_stats(date_to?: string, date_from?: string, project_id: string) — Get assistant query counts broken down by caller type (web, api, other, total) for a date range. Returns a single aggregate object; no pagination.
- mintlify_get_assistant_conversations(limit?: integer, cursor?: string, date_to?: string, date_from?: string, project_id: string) — Export the AI assistant conversation history (question, answer, cited sources, resolution status) for a Mintlify project, optionally filtered by date. Returns one page of conversations plus a cursor t
- mintlify_get_feedback(limit?: integer, cursor?: string, source?: string, status?: string, date_to?: string, date_from?: string, project_id: string) — Export a project's user feedback (thumbs / contextual comments / agent feedback), optionally filtered by date, source and status. Returns one page plus a cursor to fetch the next.
- mintlify_get_feedback_by_page(limit?: integer, source?: string, status?: string, date_to?: string, date_from?: string, project_id: string) — Get feedback counts (thumbs up/down, code, total) aggregated per documentation page path for a date range. Single page of results — this endpoint exposes no pagination cursor, so only the first page i
- mintlify_get_page_views(limit?: integer, offset?: integer, date_to?: string, date_from?: string, project_id: string) — Export per-path and site-wide content view counts split by human vs AI traffic. Offset-paginated: returns one page of rows plus site-wide totals, the hasMore flag as has_more, and a next_offset to fet
- mintlify_get_search_queries(limit?: integer, cursor?: string, date_to?: string, date_from?: string, project_id: string) — Export documentation search terms ordered by hit count (with CTR and top clicked page). Returns one page plus a cursor to fetch the next; there is no more-flag, so stop paging once next_cursor is null
- mintlify_get_unique_visitors(limit?: integer, offset?: integer, date_to?: string, date_from?: string, project_id: string) — Export per-path and site-wide approximate distinct visitor counts split by human vs AI traffic. Offset-paginated: returns one page plus a `has_more` flag and the effective `next_offset` to fetch the f
- mintlify_get_update_status(status_id: string) — Get the status and details (queued / in_progress / success / failure, logs, commit, screenshot) of a triggered deployment update by its statusId. Poll this with the statusId returned by MINTLIFY_TRIGG
- mintlify_send_agent_message(job_id: string, prompt: string, project_id: string) — Send a follow-up instruction to an existing, in-progress docs agent job (asynchronous). Use to refine or continue work on a job created by create_agent_job. Returns the updated AgentJob; poll get_agen
- mintlify_trigger_automation(project_id: string, workflow_schema_id: string) — Trigger a custom-schedule automation (workflow) to run immediately. Returns the schemaId, instanceId and jobId of the launched run. Only custom-schedule automations are triggerable (others return 400)
- mintlify_trigger_preview(branch: string, project_id: string) — Create or update a preview deployment of the docs project for a Git branch (redeploys if a preview already exists for that branch). Returns a statusId to poll with MINTLIFY_GET_UPDATE_STATUS and a pre
- mintlify_trigger_update(project_id: string) — Queue a production deployment of the docs project from its configured branch. Returns a statusId to poll with MINTLIFY_GET_UPDATE_STATUS. Requires a Mintlify plan that includes deployments (otherwise 

## Example prompts
- "What can I do in Mintlify?"
- "Show me a summary of my Mintlify account"

## Links
Docs: https://mcp.ai/docs/mcps/mintlify
Website: https://mcp.ai/mcps/mintlify