developer-toolsoauth2

Mintlify MCP

Search and retrieve documentation from Mintlify projects.

Verdict

The Mintlify MCP connects your team's AI to Mintlify's documentation platform, letting you query, update, and manage technical docs directly from Switchy Projects. Engineers and technical writers @mention it to search across doc sites, retrieve specific pages, check for broken links, or draft new content that matches your style guide. Because it uses OAuth, setup takes a few clicks but you'll need admin access to your Mintlify workspace. Most useful for teams maintaining developer docs who want AI assistance without leaving their workflow.

Common use cases

  • Search docs for outdated API examples
  • Draft changelog entries from commit messages
  • Audit documentation for broken links
  • Generate new guide pages from code comments
  • Compare docs across product versions

Integration

Vendor
Mintlify MCP
Category
developer-tools
Auth
OAUTH2
Composio slug
mintlify_mcp

Tools

Per-tool listings haven't synced yet for Mintlify MCP. The connection itself works - your Project can already @-mention it. Tool descriptions will fill in on the next Composio ingest.

Setup

Setup guide

  1. 11. In Switchy, open Settings and navigate to Integrations, then click Connect next to Mintlify MCP. 2. You'll be redirected to Mintlify's OAuth consent screen—sign in with an account that has admin permissions on your documentation workspace. 3. Grant read and write access to your docs, which allows the MCP to search content and make edits on your behalf. 4. After authorizing, Switchy confirms the connection and you'll see Mintlify MCP listed as active. 5. Open any Project and type '@Mintlify' followed by a request like 'show me all pages tagged API reference' to verify the connection works. 6. If the MCP returns results, you're ready—if not, check that your Mintlify account has the correct workspace selected and re-authorize if needed.

What teammates see: by default, memories from Mintlify MCP are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.

Works well with

Top models

Compatibility data appears once enough Projects have used this MCP together with a given model.

How Switchy teams use it

Not enough Projects yet to publish anonymised usage stats (we require ≥ 50 Projects per week).

Starter prompts

Draft API Changelog

@Mintlify review changes to API documentation in the last 30 days and draft a changelog entry summarizing new endpoints, deprecated methods, and updated parameters.

Find Outdated Examples

@Mintlify search all code examples for references to version 1.x of our SDK and list the pages where they appear so we can update them.

Generate Quickstart Guide

@Mintlify extract installation and authentication steps from our API docs and draft a quickstart guide that walks a new developer through their first request.

Compare Doc Versions

@Mintlify compare the v2 and v3 documentation sections and list any features or endpoints that exist in v2 but are missing from v3 docs.

Example outputs

Illustrative - representative of the model's voice and quality, not literal recordings.

Prompt

@Mintlify search our API docs for authentication examples and return the relevant pages

Output

Found 3 pages matching 'authentication examples' in your documentation: 1. **API Authentication Guide** (/api-reference/auth) — covers OAuth 2.0 flows and API key setup 2. **Quick Start: First API Call** (/quickstart) — includes a basic auth header example 3. **Security Best Practices** (/guides/security) — discusses token rotation and scope management Each page includes code snippets in JavaScript, Python, and cURL. The Authentication Guide has the most comprehensive coverage with 8 examples across different grant types.

Notes

This example shows doc search across your Mintlify-hosted documentation. Results depend on what's indexed in your account. OAuth scope must include read access to documentation content. Search quality improves with well-structured page titles and metadata.

Prompt

@Mintlify create a new documentation page at /api-reference/webhooks with a title 'Webhook Events' and a basic structure for event types

Output

Created new page at `/api-reference/webhooks`: **Title:** Webhook Events **Status:** Draft **Structure added:** - Overview section - Event Types table (empty, ready for content) - Payload Examples section - Retry Logic section The page is now visible in your Mintlify dashboard under API Reference. You'll need to populate the event types table and add code examples before publishing. Auto-generated navigation link added to the sidebar.

Notes

This demonstrates page creation with scaffolding. The MCP can initialize structure but won't write full content — you still author the details. Requires write permissions in OAuth scope. Changes go live immediately if your docs are set to auto-deploy; otherwise they remain in draft until you publish.

Prompt

@Mintlify review all pages tagged 'deprecated' and summarize which endpoints they document so we can plan removal

Output

Found 5 pages tagged 'deprecated' across your documentation: **REST API v1 endpoints (3 pages):** - `/v1/users` — replaced by `/v2/users` in March 2024 - `/v1/projects/list` — now `/v2/projects` with pagination - `/v1/auth/token` — superseded by OAuth 2.0 flow **Legacy webhooks (2 pages):** - Old webhook format (JSON structure changed in v2) - Deprecated event types: `project.archived`, `user.suspended` All pages show deprecation notices but are still publicly accessible. Consider adding sunset dates and migration guides before removal.

Notes

This example pairs the MCP's metadata access with AI reasoning to synthesize findings. Useful for documentation audits and technical debt planning. The AI interprets tags and page content to provide strategic context. Accuracy depends on consistent tagging practices in your Mintlify setup.

Use-case deep-dives

Code review discussions referencing API contracts

Linking PR comments to canonical method definitions

A 5-person backend team reviews a PR that changes a public API endpoint. A reviewer asks "does this break the contract we documented"—the Mintlify MCP fetches the endpoint spec from the team's API reference and pastes it into the Switchy thread. This beats tab-switching to the docs site and keeps the discussion anchored to the source of truth. The catch: if your API docs are auto-generated from OpenAPI and live in Swagger UI, this MCP won't reach them. It's also redundant if your team uses inline docstrings and the reviewer can just read the code. The sweet spot is teams with hand-written API guides in Mintlify who want to reference them without leaving the review flow.

Frequently asked

What does the Mintlify MCP do in Switchy?

It connects your Mintlify documentation workspace to Switchy so your AI agents can read, search, and reference your technical docs during conversations. Your team can ask questions about API endpoints, integration guides, or configuration steps without leaving the chat. The MCP pulls content directly from your Mintlify project, keeping answers current with your published documentation.

Do I need admin access to connect Mintlify?

Yes, you'll need owner or admin permissions on your Mintlify workspace to authorize the OAuth connection. The integration requests read access to your documentation content and project metadata. Regular contributors can't complete the setup because Mintlify's OAuth flow requires workspace-level permissions to grant API access to third-party tools like Switchy.

Can the Mintlify MCP edit or publish documentation?

No, it's read-only. The MCP retrieves published documentation content but can't create pages, update markdown files, or trigger deployments. If your team needs to modify docs based on AI suggestions, someone still opens Mintlify's editor or pushes to your Git repository manually. This keeps your documentation workflow controlled and prevents accidental overwrites during AI conversations.

How is this different from just searching Mintlify directly?

Your AI agents can cross-reference documentation with other connected tools in the same conversation. Ask a question about your API, and the agent pulls from Mintlify docs, checks your GitHub issues, and queries your database schema simultaneously. You're not switching tabs or copying snippets between tools. The context stays unified, which matters when debugging or onboarding new engineers.

Who on the team should connect this integration?

Your technical lead or whoever manages your Mintlify workspace should handle the initial OAuth setup. Once connected, any Switchy team member can use the MCP in their conversations without needing separate Mintlify credentials. The connection is workspace-wide, so you authorize once and the entire team benefits from documentation access in their AI chats.

More developer tools MCPs

See all developer tools MCPs →

Data last verified 61 hours ago.Sources aggregated hourly to weekly. See docs/architecture/model-directory.md.