writers-room
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- NO_CODE (SAFE): The skill contains only Markdown documentation and prompt instructions. No scripts (Python, JavaScript, or Shell) are included in the repository.
- DATA_EXFILTRATION (SAFE): No network activity (curl, wget, fetch), sensitive file access (SSH keys, AWS configs), or hardcoded secrets were detected.
- INDIRECT_PROMPT_INJECTION (SAFE): The skill processes user-provided content (e.g., 'Existing material' in SKILL.md). Mandatory Evidence Chain: 1. Ingestion point: SKILL.md 'Brief the Room' section. 2. Boundary markers: None present. 3. Capability inventory: None detected (no code execution or network ops). 4. Sanitization: None present. Due to the complete lack of any executable capabilities, this ingestion surface is classified as safe.
Audit Metadata