Use the hosted MCP interface powered by the Repokit API

How MCP fits the product

The Repokit API is the direct engine boundary for repository-aware retrieval.

Repokit MCP is the first-party hosted MCP interface for clients that prefer remote tools over direct HTTP integration.

Endpoints

• Host root: https://mcp.repokit.live
• HTTP MCP endpoint: https://mcp.repokit.live/mcp

Where to start reading

Use the human-facing MCP page at https://repokit.co/mcp when you want the product overview.

Use the host root and /mcp endpoint below only when your client actually needs them. Use the API page instead if you want direct HTTP integration.

Authentication

Initialize can be unauthenticated, but authenticated tool calls should use the verification token issued for your ready repository.

Initialize example

curl -sS -X POST \
  -H "Content-Type: application/json" \
  https://mcp.repokit.live/mcp \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "initialize",
    "params": {
      "protocolVersion": "2025-03-26",
      "capabilities": {},
      "clientInfo": {
        "name": "repo-intelligence-docs-example",
        "version": "1.0.0"
      }
    }
  }'

Tools list example

curl -sS -X POST \
  -H "Content-Type: application/json" \
  https://mcp.repokit.live/mcp \
  -d '{
    "jsonrpc": "2.0",
    "id": 2,
    "method": "tools/list",
    "params": {}
  }'

Tool call example

curl -sS -X POST \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <your_token>" \
  https://mcp.repokit.live/mcp \
  -d '{
    "jsonrpc": "2.0",
    "id": 3,
    "method": "tools/call",
    "params": {
      "name": "find_relevant_files",
      "arguments": {
        "repository_id": "<repository_id>",
        "query": "fix request validation error in routing",
        "top_k": 3
      }
    }
  }'

When to use MCP instead of the API

• Your client already supports MCP.
• You want tool-style access rather than direct endpoint wiring.
• You want a simpler agent integration path while keeping the same underlying retrieval engine.

Related docs