skills/github/gh-aw/copilot-review/Gen Agent Trust Hub

copilot-review

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes user-controlled data from GitHub pull request comments, creating an attack surface for indirect prompt injection.\n
  • Ingestion points: Pull request content and comments are retrieved via gh pr view in SKILL.md.\n
  • Boundary markers: The skill includes logic to filter authors by association (MEMBER, OWNER, COLLABORATOR) and specific bot IDs (e.g., github-actions[bot]).\n
  • Capability inventory: The agent is instructed to modify code and resolve review threads based on the feedback received.\n
  • Sanitization: While author-based filtering is used, the content of the comments is not specifically sanitized or escaped before the agent processes them for code changes.\n- [COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) to fetch pull request data and reviews. This is a standard vendor-provided tool used for the skill's intended functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 07:07 PM
Security Audit — agent-trust-hub — copilot-review