claude-md-authoring
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- NO_CODE (SAFE): The skill consists entirely of markdown documentation and guidelines. No scripts (Python, Bash, Node.js) or executable files are included in the package.
- EXTERNAL_DOWNLOADS (SAFE): No external URLs or remote resources are referenced for download or execution.
- CREDENTIALS_UNSAFE (SAFE): No hardcoded secrets, API keys, or sensitive environment variables were detected.
- DATA_EXFILTRATION (SAFE): The skill does not perform any network operations or file system reads that would lead to data exposure.
- PROMPT_INJECTION (SAFE): No prompt injection patterns or attempts to bypass AI safety filters were found. The instructions are focused on guiding the user or agent in writing structured configuration documentation.
Audit Metadata