Skip to main content

Docs MCP endpoint

The documentation site exposes an MCP server at:
This endpoint is provided by Mintlify and serves the published documentation content. It supports the Model Context Protocol over HTTP.

Direct URL (streamable HTTP)

Clients that accept an MCP server URL directly:

Command-based (stdio via mcp-remote)

Clients that require a command and arguments:

Setup by client

  1. Open Settings then Connectors.
  2. Click Add custom connector.
  3. Set the name to “WebMCP Docs” and the URL to https://docs.mcp-b.ai/mcp.
  4. Click Save, then Connect.
Add to .cursor/mcp.json in your project root:
Restart Cursor after saving.
Add to mcp_config.json:
Restart Windsurf after saving.
MCP support in ChatGPT requires a paid plan with Developer Mode enabled.
  1. Enable Developer Mode in ChatGPT settings.
  2. Go to Settings, then Connectors.
  3. Click Add MCP server.
  4. Enter https://docs.mcp-b.ai/mcp as the server URL.
  5. Name it “WebMCP Docs” and click Save.
Add to ~/.codex/config.toml:
Restart Codex after saving.

What the server provides

The MCP server gives connected agents read access to the full documentation site. Agents can search for pages, retrieve page content, and browse the site structure.

llms.txt

Mintlify automatically generates and hosts two files at the root of the documentation site. These files follow the llmstxt.org convention for AI-readable site indexing.

Files

How they are generated

Both files are derived from docs.json and the published .mdx pages. Mintlify rebuilds them on every deployment. There is no manual authoring step. The llms.txt file includes:
  • A top-level description (from the description field in docs.json)
  • A quick start code snippet
  • A package summary table
  • A flat sitemap of every page with its title and relative URL
The llms-full.txt file contains the full markdown-rendered body of every page, separated by page headers.

Maintenance

Because these files are auto-generated, the only way to change their content is to update the source pages or docs.json. You do not edit llms.txt or llms-full.txt directly.

Usage

AI assistants (Claude, ChatGPT, Gemini, Perplexity, and others) fetch llms.txt to understand what the site covers before retrieving specific pages. If your agent supports URL-based context, point it at https://docs.mcp-b.ai/llms.txt. The MCP endpoint and llms.txt serve different purposes. llms.txt is a static file that agents fetch over plain HTTP. The MCP endpoint is an interactive server that agents connect to as an MCP client, allowing structured queries against the documentation.

Versioning and changelog

WebMCP uses Changesets for versioning. Each package in the @mcp-b/* scope is versioned independently and published to npm.

Where to find changelogs

Versioning policy

All packages follow Semantic Versioning:

Package scope

The monorepo publishes the following packages. Each has its own version number.

Reporting issues

If you find a bug or regression after upgrading:
  1. Check existing issues.
  2. Open a new issue with the package name, version numbers, steps to reproduce, and expected versus actual behavior.