Skip to main content
Quantora runs a hosted MCP server, so an agent in Claude, Claude Code, Cursor or any other MCP client can query the same data the REST API serves. There is nothing to install and no Node process to keep running: point the client at the endpoint and authenticate with your API key.
Every tool is read-only. The server reads data, it cannot change your account, your keys or your billing.

Endpoint

Authentication uses the same key as the REST API, sent either as an x-api-key header or as Authorization: Bearer YOUR_API_KEY. Create and manage keys in your account console.
Authentication is a static API key, there is no OAuth flow yet. Your key ends up in a client config file on disk, so treat it like any other secret and rotate it from the console if a machine is shared or lost.

Connect your client

1

Claude Code

Run this in your terminal, with your key in place of YOUR_API_KEY:
2

Claude Desktop or Cursor

Add the server to your MCP config (claude_desktop_config.json, or Cursor’s mcp.json), then restart the app:
3

Clients without HTTP transport

Older clients that only speak stdio can bridge through mcp-remote:

Tools

The server exposes 33 tools. They return the same objects, units and trust metadata as the REST endpoints they sit on, so anything you read in Core concepts applies here too.

Discovery and resolution

Macro observations

Screeners

Companies

Markets, FX and positioning

Events and provenance

Rights and freshness still apply

The MCP tools are a different transport, not a different dataset. Two fields decide whether a row is safe for what you are building, and both come back through MCP exactly as they do over HTTP:

Sources and rights

Which series may be redistributed, and which are licensed for internal use only.

Trust metadata

Freshness status and scoring eligibility, so an agent can tell a stale row from a current one.