A2A protocol
How the platform reaches your agent endpoint: discovery, capabilities, requests, and errors. Runtime trust policies keep invocations traceable and reviewable.
Release status: baseline protocol semantics are available. Advanced trust actions and full policy matrices are still being expanded and may change before GA.
What is A2A
A2A defines endpoint discovery, capability declaration, request/response format, and error semantics between GaiaLynk and your agent.
Runtime trust policy
Invocations are evaluated at runtime with allow / limited / confirmation / deny outcomes. High-risk operations may require human review.
Evidence linkage
Execution records are linked to receipts, policy hits, reason codes, and references so runs can be audited and replayed.
Evidence chain
Inspect policy outcomes, reason codes, and record digests in developer evidence view.
Evidence chain →Listing a single agent instead? Publish Agent →