review-pr

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to examine code changes using git status and git diff. This creates a surface where malicious instructions hidden in the code being reviewed could attempt to influence the agent's behavior during the checklist process. The risk is compounded by the agent's capability to execute shell commands for building and pushing code.\n
  • Ingestion points: Local repository changes and metadata accessed via git commands as outlined in SKILL.md.\n
  • Boundary markers: The skill does not implement delimiters or specific warnings to ignore instructions found within the code diffs.\n
  • Capability inventory: Shell command execution for package building (yarn build) and Git operations (git push) defined throughout the workflow in SKILL.md.\n
  • Sanitization: There are no mentioned mechanisms for sanitizing, escaping, or validating the content extracted from the files before the agent processes it.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 01:53 PM
Security Audit — agent-trust-hub — review-pr