mece-writing
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is strictly instructional, providing a logical workflow for structured writing. No obfuscation, prompt bypass attempts, or hardcoded credentials were found.- [COMMAND_EXECUTION]: The skill directs the agent to save output to local files using a specific naming convention: 'mece-[keyword]-[YYYYMMDD].md'. This structured approach limits the risk of arbitrary file creation or overwriting sensitive system files, as the scope is restricted to the current working directory and the markdown format.- [PROMPT_INJECTION]: While the skill processes user-supplied topics (an indirect injection surface), it employs a multi-step verification workflow (Step 3: MECE Validation) that requires the agent to logically validate the structure before writing. This acts as a functional guardrail against malicious input intended to hijack the agent's behavior.
Audit Metadata