document-it

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes repository code and existing documentation to generate updates, which establishes an inherent surface for indirect prompt injection attacks.
  • Ingestion points: Local repository files, source code, and DocSlime documentation files (SKILL.md).
  • Boundary markers: No specific delimiters or safety headers are defined to distinguish ingested content from the agent's instructions.
  • Capability inventory: The skill performs file system modifications and executes internal CLI tools for documentation management (SKILL.md).
  • Sanitization: No explicit instructions are provided for sanitizing or filtering external data before processing.
  • [SAFE]: The instructions provide positive security guidance, specifically directing the agent to ensure 'No secrets' are added to the documentation and to respect all host-level permission prompts for risky commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 01:31 AM