self-review

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists entirely of natural language instructions designed to guide the agent through a code quality review process.
  • [DATA_EXPOSURE]: The skill instructs the agent to read changed files in the project. This is standard behavior for a development-oriented skill and does not involve exfiltration to external domains.
  • [COMMAND_EXECUTION]: The process section includes a step to fix identified issues, which utilizes the agent's existing file-writing capabilities to perform the intended self-correction task.
  • [PROMPT_INJECTION]: The inclusion of the $ARGUMENTS placeholder at the end of the file is a standard mechanism for passing parameters and does not contain any malicious override instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 03:06 PM
Security Audit — agent-trust-hub — self-review