docs-writer
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill requests user confirmation to execute a local formatting command (npm run format) to ensure project consistency.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by auditing repository files.
- Ingestion points: Reads from the /docs directory, packages/ directory, and any .md files.
- Boundary markers: No explicit delimiters or instructions to ignore instructions found within audited data.
- Capability inventory: Can modify local files and execute the project formatting command (SKILL.md).
- Sanitization: Not present; the skill relies on technical writing standards to filter content.
Audit Metadata