Skills
skills/artblue/ai-toolkit/universal-context-auditor/Gen Agent Trust Hub

universal-context-auditor

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: No malicious activities or patterns were identified. The skill acts as an advisor for structuring AI instructions.
  • [PROMPT_INJECTION]: The skill processes untrusted user-provided content in the form of AI configuration files. 1. Ingestion points: Configuration files such as CLAUDE.md and .cursorrules are ingested via the Read tool. 2. Boundary markers: The instructions recommend using XML tags in audited files to improve structural clarity. 3. Capability inventory: Uses Read and Glob tools to acces s file content. 4. Sanitization: Content is analyze d for metrics and structural patterns; no direct execution of ingested content occurs.
  • [EXTERNAL_DOWNLOADS]: The documentation provides instructions that involve download ing content from the author's GitHub repository ('ArtBlue/AI-toolkit').
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 04:59 PM