Dominion Observatory API
API by Dominion Observatory · rest
Runtime behavioral trust scoring for the MCP agent economy. Look up indexed MCP projects; only measured endpoints have trust scores; UNRATED means not measured, report tool call telemetry, detect anomalies, and generate compliance reports. A score is evidence to review, not a safety guarantee; UNRATED means no runtime measurement. Decision thresholds: score >= 60 → PASS (proceed), 40-59 → UNCERTAIN (caution), < 40 → FAIL (block). Privacy: only server_url, success, latency_ms, tool_name, http_status collected. No prompts, arguments, outputs, user IDs, or IPs.
Endpoint: https://dominionobservatory.com/
Verification: observed. Source: crawl.
Capabilities (27)
- Ed25519 public keys for signed captures (distinct from attestation JWKS)
- A2A JSON-RPC interface descriptor
- A2A JSON-RPC request
- Payment-gated full trust verdict (x402)
- Scan a server for agent-readiness
- SVG trust score badge
- Current product capabilities and explicit limitations
- Compliance audit export
- Aggregate pre-flight vetting demand across the trust network
- Tool-manifest drift check for an MCP server (free)
- Capture and publish a public endpoint manifest then compare with the previous…
- Top servers by trust score
- NSA MCP compliance checklist for a server
- x402 payment configuration
- Report a tool call outcome
- SLA tier certification
- Observatory statistics
- Get behavioral trust score for a server
- Trust score delta feed
- Full server benchmark report
- A2A behavioral evidence attestation
- Standards compliance attestation (CI/CD format)
- ERC-8004 endpoint health attestation
- Pre-flight due-diligence check (call before invoking a server)
- Resolve a diligence-receipt id to the full signed claim