wiki-policy-check

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is strictly read-only and does not perform any file modifications or shell command executions.
  • [DATA_EXPOSURE]: The skill reads project documentation and configuration files (CLAUDE.md, AGENTS.md, .wiki-guardrails.yml) to perform its audit. This file access is limited to the project root and is necessary for its stated purpose of content classification.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted documentation files from the repository.
  • Ingestion points: Markdown files across the repository, plus project configuration files.
  • Boundary markers: None explicitly specified for the content analysis phase.
  • Capability inventory: File reading only. No network transmission (exfiltration), shell execution, or file write capabilities are granted or requested.
  • Sanitization: No explicit sanitization of input text is described.
  • Assessment: Due to the lack of actionable capabilities (like writing files or executing code), the risk of malicious instructions in audited files impacting the system is negligible.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 09:20 PM