add-review-comment

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: Uses the standard gh (GitHub CLI) tool to interact with pull requests. The skill instructions use subcommands like gh api and gh pr view to post comments and resolve commit SHAs, which is expected behavior for this use case.
  • [DATA_EXFILTRATION]: Communicates with GitHub's REST API. GitHub is a well-known service, and the operations performed (posting comments) are consistent with the skill's stated purpose. There is no evidence of sensitive data harvesting or unauthorized network connections.
  • [SAFE]: The skill implements a form of least-privilege by specifying restricted allowed-tools (e.g., gh api:*, gh pr view:*), which limits the agent's execution scope to the necessary GitHub subcommands.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 08:50 AM
Security Audit — agent-trust-hub — add-review-comment