docu-optimizer

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
  • Ingestion points: The skill reads untrusted content from CLAUDE.md, the docs/ directory, and project source files (to extract APIs and project stages).
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when passing documentation content to the five subagents defined in Phase 2.
  • Capability inventory: The skill utilizes tools such as Edit, Write, and Bash, which could be triggered maliciously if the agent obeys instructions hidden within documentation or code comments.
  • Sanitization: No sanitization or structural validation is mentioned for the content extracted from the repository before it is synthesized into a final report or an 'optimized' file version.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 02:01 PM