MCP SERVER

Your docs, inside every AI coding tool

Every BeforeQuery knowledge base doubles as a Model Context Protocol server. Cursor, Claude, and any MCP client can search and query your current documentation — so AI assistants answer from your real docs instead of stale training data.

Start FreeAPI Docs

A standards-based endpoint

MCP and A2A, no extra infrastructure

Three tools, JSON-RPC

MCP Tools

  • search_docs — hybrid search over your knowledge base
  • ask_question — full RAG answers with citations
  • list_sources — enumerate connected sources
  • One endpoint per knowledge base
  • Standard JSON-RPC, works with any MCP client
No stale answers

Always Current

  • Backed by the live index, not a snapshot
  • GitHub/GitLab webhook re-sync on every push
  • Website sources re-crawled automatically
  • Deprecated endpoints stop being recommended
  • Same retrieval quality as the widget and API
Keys, not open doors

Access Control

  • Authenticated with client keys you issue
  • Scoped per knowledge base
  • Revoke a key to cut access instantly
  • Requests observable in traces
  • A2A AgentCard for agent-to-agent interop
Why it matters

Developers ask their editor first now

The first place a developer asks about your API is no longer your docs site — it's the AI assistant in their editor. Without MCP, that assistant answers from training data frozen months ago. With your knowledge base exposed as an MCP server, it retrieves your current documentation and answers with real citations.

  • Works in Cursor, Claude, and any MCP-compatible client
  • Answers reflect today's docs, including this morning's push
  • A2A endpoint lets other agents query your knowledge base
  • Agents SDK (@beforequery/agents) includes MCP and A2A clients
  • Traces record every MCP call for observability
POST /mcp/knowledge-bases/:id
{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "name": "ask_question",
    "arguments": {
      "question": "How do I paginate the list endpoint?"
    }
  },
  "id": 1
}

Frequently Asked Questions

Common questions about MCP Server

Any MCP-compatible client — Cursor, Claude, and the growing ecosystem of MCP tools. The server speaks standard JSON-RPC with three tools: search_docs, ask_question, and list_sources.
Each knowledge base's MCP endpoint is authenticated with client keys you create and revoke from the dashboard. Keys are scoped per knowledge base, and every call is recorded in traces.
No. The MCP server is built into BeforeQuery — every knowledge base gets an endpoint at /mcp/knowledge-bases/:id automatically. Configure a client key in your MCP client and you're done.
A2A (agent-to-agent) is a companion protocol: each knowledge base publishes a public AgentCard and accepts message/send calls, so other AI agents can query your knowledge base programmatically. If you only want in-IDE answers, MCP alone is enough.
As current as the index. Repo sources re-sync via webhook on every push, websites re-crawl on schedule, and MCP queries hit the live index — there is no separate export or snapshot to maintain.

Expose your docs over MCP

Create a knowledge base, issue a client key, and your docs are in the IDE.

Get Started Free