Free, no key, CORS-open, rate-limited at 60 requests/minute and 500 requests/day per source. Grades refresh continuously (full probe cycle ≈ 3 days). Data is CC BY 4.0 — use it freely, with attribution and a link to mcpqueen.com. Every response carries its own attribution, license, and methodology fields.
| GET /api/grades.json | Top 500 graded servers by score — grade, score, provisional flag, latency, tool count, protocol-access state, probe time. Protocol access describes handshake/discovery only, not data entitlement. |
| GET /api/history/{name}.json | Per-server probe time series (last 200 probes). Example: /api/history/com.healthai/clarity.json |
| GET /api/changes.json | Latest 100 grade transitions across the registry — who got better, who broke. |
| GET /api/trust/{name}.json | Per-server trust receipt: protocol grade, separate security/data/citation/claim observations, and reviewed field reports. |
| GET /badge/{name}.svg | Live grade badge for a server, e.g. /badge/com.healthai/clarity.svg — embed it in a README. |
The registry is itself an MCP server. Search capabilities with search_servers/search_tools, then inspect evidence with get_trust_receipt/search_trust_evidence.
claude mcp add --transport http mcpqueen https://mcpqueen.com/mcp
Or raw JSON-RPC:
curl -X POST https://mcpqueen.com/mcp \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call",
"params":{"name":"search_servers","arguments":{"query":"postgres"}}}'
Details and etiquette: For Agents · machine summary: /llms.txt
Want webhooks, full history exports, or bulk access? That tier is coming — the data already exists. Watch this page.