merge-pr

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill leverages several command-line utilities to automate developer workflows, including gh for GitHub interactions, cargo for Rust project tasks, and bun for documentation processing. These tools are used as intended for viewing PRs, checking CI status, and merging code.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it reads untrusted content from pull request titles, descriptions, and diffs using the gh pr view and gh pr diff commands.
  • Ingestion points: Pull request content is ingested during the 'Resolve PR' and 'Metadata reconcile' steps in SKILL.md.
  • Boundary markers: There are no explicit markers or instructions provided to separate the external PR data from the agent's internal logic.
  • Capability inventory: The skill possesses the capability to merge pull requests via gh pr merge and modify the repository's filesystem (specifically documentation and roadmap files).
  • Sanitization: The instructions do not specify any sanitization or validation of the PR data before it is acted upon.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 11:14 AM
Security Audit — agent-trust-hub — merge-pr