verification-before-merge

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill is susceptible to indirect prompt injection through data ingestion points from GitHub.
  • Ingestion points: The agent reads the issue body (gh issue view --json body), PR reviews (gh pr view --json reviews), and review threads (gh pr view --json reviewThreads).
  • Boundary markers: There are no explicit instructions or delimiters telling the agent to treat the issue body or PR comments as untrusted data rather than instructions.
  • Capability inventory: The skill possesses high-impact capabilities including pushing code (git push --force-with-lease), merging PRs (gh pr merge), and executing project-defined scripts (pnpm test).
  • Sanitization: There is no evidence of sanitization or filtering of the content retrieved from GitHub.
  • [Command Execution] (SAFE): The skill executes several shell commands using standard tools (git, gh, pnpm, docker-compose).
  • These are necessary for the skill's primary purpose of PR verification and merging.
  • The use of git push --force-with-lease is a high-impact operation but is appropriate within the context of a rebase-based workflow described in the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:32 PM