meta-reviewing-web-reviewing

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious execution patterns, hardcoded credentials, or unauthorized network operations were detected. The skill uses standard markdown and instructional text to define code review patterns.\n- [PROMPT_INJECTION]: The skill is designed to process untrusted external data in the form of React component diffs and JSX/TSX files, which introduces a surface for indirect prompt injection.\n
  • Ingestion points: React component source code and PR diffs provided by users for review in files such as SKILL.md.\n
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the code being reviewed.\n
  • Capability inventory: The skill provides instructions for the agent to analyze code structure, logic, and accessibility patterns.\n
  • Sanitization: The skill does not implement explicit sanitization or filtering of the input code, relying instead on the agent's baseline safety protocols.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 07:37 AM
Security Audit — agent-trust-hub — meta-reviewing-web-reviewing