pattern-critic
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFENO_CODE
Full Analysis
- Indirect Prompt Injection (SAFE): The skill is designed to ingest untrusted code files for review. While this constitutes an ingestion point for potentially malicious instructions embedded in comments or code, the skill has no associated capabilities (e.g., subprocess execution, network requests, or file-system writing) to exploit such an injection. The impact is limited to the potential for the agent to deviate from its review task.\n
- Ingestion points:
SKILL.md(Workflow step 1: 'Read the pattern file to review').\n - Boundary markers: None present in the instructions.\n
- Capability inventory: None detected; the skill only generates text output.\n
- Sanitization: None detected.\n- No Code (SAFE): The skill is composed exclusively of metadata and documentation in a markdown file. It does not include Python scripts, Node.js packages, or any form of executable code, which eliminates common vectors for remote code execution or system compromise.
Audit Metadata