release-pr-gates

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill processes potentially untrusted content from GitHub Actions logs, commit histories, and pull request metadata. It implements robust defenses by instructing the agent to treat this content strictly as evidence and to ignore any instructions embedded within the data.
  • Ingestion points: Data enters the context via gh run view --log, git log, and gh pr list commands in SKILL.md.
  • Boundary markers: Explicit instructions are provided in the "External Side Effects" section of SKILL.md to disregard embedded commands and treat text as untrusted.
  • Capability inventory: The agent has permissions to execute shell commands, manage repository tags, and create GitHub releases.
  • Sanitization: The instructions specifically require the redaction of secret-like values from external data before processing.
  • [COMMAND_EXECUTION]: The skill uses npm, bun, and bunx to execute local build, formatting, and linting scripts defined in the repository. This is an expected and necessary function for verifying release readiness and occurs within the local environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 01:29 PM
Security Audit — agent-trust-hub — release-pr-gates