docgen
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is configured to run local project scripts, specifically 'node scripts/generate-usage-docs.mjs' and './scripts/check-usage-docs-sync.sh', to generate documentation and verify its status.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it reads and processes project source files to extract documentation metadata. 1. Ingestion points: Scans directories including 'skills/', 'cli/src/', and 'scripts/' as specified in 'SKILL.md' and 'guide.md'. 2. Boundary markers: No explicit delimitation or instructions are provided to the agent to disregard potential instructions embedded in the analyzed source code. 3. Capability inventory: The skill has access to powerful tools including 'Bash(*)', 'Write', and 'Read'. 4. Sanitization: There is no evidence of filtering or sanitization of content extracted from source code comments or metadata before it is processed.
- [SAFE]: The skill's primary operations and requested tool access align with its stated purpose of project documentation generation, using local vendor-provided scripts and resources from the same organization (AsiaOstrich).
Audit Metadata