NYC

documentation-and-adrs

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [SAFE] (SAFE): No malicious patterns, executable code, or obfuscation were detected in the skill content.
  • [Indirect Prompt Injection] (LOW): The skill encourages the use of project documentation (ADRs, CLAUDE.md) as high-priority context for AI agents, creating an ingestion surface for external data. * Ingestion points: docs/decisions/, README.md, and CLAUDE.md. * Boundary markers: Uses standard Markdown headers but lacks explicit instructions to ignore embedded commands. * Capability inventory: The skill contains no code or active capabilities. * Sanitization: None; documentation is treated as trusted context.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:38 PM