release-health-gates

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by directing the agent to ingest and act upon data from external GitHub resources.
  • Ingestion points: Data enters the agent's context through GitHub API calls (check-suites, deployments) and the monitoring of issue labels (e.g., 'qa-blocker') as specified in 'modules/quality-signals.md'.
  • Boundary markers: The instructions lack explicit boundary markers or directives to treat external data as untrusted, which could lead the agent to interpret embedded comments or issue descriptions as legitimate instructions.
  • Capability inventory: The skill utilizes the 'minister-tracker' tool and generates markdown snippets for PR comments, providing a mechanism for poisoned input to propagate into repository artifacts or influence agent decisions.
  • Sanitization: No sanitization, validation, or escaping of the ingested GitHub content is described in the provided modules.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 06:51 PM