github-workflow-automation

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing untrusted data from GitHub repositories.
  • Ingestion points: Untrusted data enters the agent context through PR diffs in ai-review.yml, issue bodies in issue-triage.yml, and comment text in mention-bot.yml.
  • Boundary markers: The workflow templates do not include clear delimiters or instructions for the AI to ignore instructions embedded within the interpolated diffs or issue content.
  • Capability inventory: The skill's workflows have permissions to write pull request reviews, add labels, create issue comments, and execute git operations including git push --force-with-lease.
  • Sanitization: There is no evidence of input validation or content filtering for the data ingested from GitHub.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes official GitHub Actions (such as actions/checkout and actions/github-script) and the @anthropic-ai/sdk library, which are from well-known and reputable sources.
  • [SAFE]: Sensitive credentials such as the ANTHROPIC_API_KEY are managed using GitHub Secrets, adhering to industry security standards for CI/CD environments.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 08:55 PM
Security Audit — agent-trust-hub — github-workflow-automation