publish-pr-review

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the gh api command to interact with the GitHub API. It specifically constructs a POST request to the PR reviews endpoint using a single-quoted heredoc to safely handle the JSON payload, which is the primary intended function of the skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from pull request comments and existing reviews, creating a potential surface for indirect injection. However, this is addressed by clear safety constraints.
  • Ingestion points: The skill reads existing PR reviews, threads, and comments as stated in SKILL.md.
  • Boundary markers: A mandatory human-in-the-loop confirmation step is enforced: 'The exact assembled review body, every inline comment, and every inline location have not been shown and confirmed. Show the copy itself as a separate step before any mutation'.
  • Capability inventory: The skill utilizes the gh CLI tool to submit data to GitHub.
  • Sanitization: The skill requires using wording previously approved by the user and suggests passing content through a prose-checking utility when available.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 01:37 PM