git-review

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local shell commands to perform its functions. It uses git to analyze repository history and uncommitted changes, and employs the GitHub CLI (gh) to retrieve issue and pull request details. Additionally, it identifies and executes project test suites using frameworks such as pytest, npm, cargo, and go to validate implementations as defined in SKILL.md phase 5.\n- [EXTERNAL_DOWNLOADS]: The skill fetches external data (issue and pull request metadata) from GitHub using the official gh CLI tool. This is a legitimate operation for a code review tool to establish the context and requirements of the changes being reviewed.\n- [PROMPT_INJECTION]: The skill processes untrusted content from GitHub issues and pull requests as requirements, representing an indirect prompt injection surface. Ingestion points: gh issue view and gh pr view commands (Phase 1b). Boundary markers: Not explicitly defined for the interpolated requirement text. Capability inventory: Shell execution of project tests (Phase 5). Sanitization: No sanitization of external text is described. Despite the presence of the surface, the skill is focused on comparing code to requirements rather than following instructions within that text.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 01:11 AM