Agent Health Monitor (AHM)

• Trust Score Evaluation trust_score_evaluation

  Evaluates an AI agent's trustworthiness based on behavioral analysis, historical performance, compliance posture, and peer attestations. Returns a composite trust score (0-100) with dimensional breakdowns.

  trust scoring evaluation risk-assessment
  3 examples

  • Evaluate the trust score for agent at https://agent.example.com/.well-known/agent.json
  • What is the current trust rating for agent id abc-123?
  • Run a full trust assessment on this agent

  In: application/json · Out: application/json

• Agent Health Monitoring agent_health_monitoring

  Continuously monitors an AI agent's operational health including uptime, response latency, error rates, and behavioral consistency. Supports push notifications for health status changes.

  health monitoring uptime observability
  3 examples

  • Monitor the health of agent https://agent.example.com
  • Get the current health status for all monitored agents
  • Set up health alerts for agent id xyz-789

  In: application/json · Out: application/json

• Verifiable Credential Issuance verifiable_credential_issuance

  Issues W3C-compliant verifiable credentials attesting to an agent's trust score, health status, or compliance with specific standards. Credentials can be cryptographically verified by any relying party.

  credentials verifiable attestation compliance W3C-VC
  3 examples

  • Issue a trust credential for agent abc-123 with current score
  • Generate a compliance attestation for agent xyz-789
  • Create a verifiable health certificate for this agent

  In: application/json · Out: application/json, application/vc+ld+json

• Trust Report Generation trust_report_generation

  Generates detailed trust assessment reports for an agent, including historical trends, dimensional analysis (reliability, safety, transparency, compliance), and comparative benchmarks.

  report analytics trust-history benchmarking
  3 examples

  • Generate a trust report for agent abc-123 covering the last 30 days
  • Compare trust scores across all monitored agents
  • Show trust score trends for my agent fleet

  In: application/json · Out: application/json, text/markdown

• Agent Registration agent_registration

  Registers a new AI agent for trust scoring and health monitoring. Accepts an agent's A2A Agent Card URL or manual configuration. Begins baseline trust assessment upon registration.

  registration onboarding discovery
  3 examples

  • Register agent at https://new-agent.example.com/.well-known/agent.json for monitoring
  • Add this agent to the AHM trust monitoring pool
  • Onboard a new agent with manual configuration

  In: application/json, text/plain · Out: application/json

Interfaces

Capabilities

Streaming

No

Push notifications

Yes

Extended card for signed-in callers

Yes

Security and formats

Requires

Not declared

Schemes

bearerAuth (unknown), apiKey (unknown)

Accepts

application/json, text/plain

Returns

application/json