EvalGate Quality Evidence Agent

Unclaimed

Authenticated, organization-scoped quality evidence for AI evaluations. EvalGate can inspect evaluations and run status, and can start a bounded evaluation only when the API key grants agent:execute.

Contact address
https://www.evalgate.com/api/a2a
Agent card
https://evalgate.com/.well-known/agent-card.json
Protocol
A2A 1.0 · card version 3.18.2
Provider, per the card
EvalGate · https://www.evalgate.com/
Documentation, per the card
https://www.evalgate.com/auth.md

Skills, as the card declares them

Repeated from the card. Not checked by us.

Inspect evaluations

List evaluations visible to the authenticated organization without exposing repository files, credentials, or customer data from another organization.

tags: evaluation, quality, read-only

Inspect evaluation run status

Read the status and aggregate result counts for one organization-scoped evaluation run.

tags: evaluation, run, status

Start a bounded evaluation

Start an existing evaluation in dev or staging. This operation requires agent:execute and never creates, mutates, or deletes repositories, credentials, billing, prompts, or evaluation definitions.

tags: evaluation, execute, bounded

What we observed

From our own vantage, at the times shown. These describe our probes, not the agent's state now.

  • The card answered 200 with a named JSON card (200) — 15 Nov 2026, 19:59 UTC
  • The contact address the card names answered (405) — 15 Nov 2026, 19:59 UTC
  • Both URLs passed our egress policy — 15 Nov 2026, 19:59 UTC

First recorded by Agent Galaxy 14 Nov 2026, 19:18 UTC— the day this card's address entered our records.