CitedNearby

MCP server

Connect AI assistants and your own code with the Model Context Protocol.

The Model Context Protocol (MCP) lets AI assistants and your own code read your account and start scans. Every answer is only about your own account.

POST

/mcp

The one MCP endpoint (Streamable HTTP). It takes JSON-RPC 2.0 messages. GET and DELETE are not supported.

Sign in with your API key as a Bearer token, or with an OAuth access token. Without one, the server answers 401 with a header that points to our sign-in server, so MCP clients can start OAuth on their own.
Send Content-Type: application/json and Accept: application/json, text/event-stream.

Server address: https://citednearby.com/mcp

List the tools

curl

curl -X POST https://citednearby.com/mcp \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -H "Accept: application/json, text/event-stream" \ -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

Call a tool: get_account

curl

JSON

{ "jsonrpc": "2.0", "id": 1, "result": { "content": [ { "type": "text", "text": "{\"name\":\"Jane Smith\",\"credits_balance\":1250, \"...\": \"...\"}" } ], "structuredContent": { "name": "Jane Smith", "credits_balance": 1250, "...": "..." }, "isError": false } }

Tool answers come back as JSON text, and also as structured content. If a tool cannot do what you asked (a wrong id, no scan permission, not enough credits), the result has isError set to true and a plain-English message.

Every tool and its parameters are on the MCP tools page. Connecting Claude, ChatGPT, Cursor or another assistant? Use the step-by-step guide.