@trustscoreagent/mcp-server

@trustscoreagent/mcp-server

MCP server for TrustScoreAgent — a free, open reputation registry for AI microservices. It lets an LLM agent check whether an external service is trustworthy before calling it, and contribute ratings afterward.

Tools

Tool	Purpose
check_reputation	Get the trust score (0–1), confidence, rating count and dimensional breakdown (availability, latency, conformity) of a service. Call it before using an untrusted service.
submit_rating	Rate a service after calling it, from your interaction metrics (status code, latency, schema validity, optional quality score and receipt).
list_services	Discover rated services sorted by trust score, with score/ratings filters.

Install

npm install -g @trustscoreagent/mcp-server

Configure

Add it to your MCP client (e.g. Claude Desktop claude_desktop_config.json):

{
  "mcpServers": {
    "trustscoreagent": {
      "command": "trustscoreagent-mcp"
    }
  }
}

Environment variables

Variable	Default	Description
TRUSTSCORE_API_URL	production API	Base URL of the TrustScoreAgent API.
TRUSTSCORE_AGENT_DID	auto-generated	Stable identifier for this agent. If unset, a unique id is generated on first run and stored in ~/.trustscoreagent/agent-id.

Develop

npm install
npm run build      # compile TypeScript to dist/
npm run dev        # run from source with tsx

License

Apache-2.0 — see LICENSE.