vibe-writing
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- NO_CODE (SAFE): The skill consists exclusively of Markdown documentation and templates. There are no scripts, binaries, or automated installers present.\n- SAFE (SAFE): A thorough review of all 10 threat categories confirms the absence of malicious patterns, prompt injections, obfuscation, or persistence mechanisms.\n- Indirect Prompt Injection (SAFE): The skill processes untrusted user input during the 'Learning' phase. However, the risk is negligible as the agent lacks network access or code execution capabilities. (1) Ingestion points: User dialogue in learning-guide.md and project files. (2) Boundary markers: Markdown card templates in card-templates.md. (3) Capability inventory: Local markdown file creation. (4) Sanitization: None; relies on the LLM's default safety guardrails.
Audit Metadata