The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown-based instructions and templates. It does not contain any code, binaries, or system-level commands that could pose a security risk.
[PROMPT_INJECTION]: The core principles and uncertainty policy require the agent to prioritize code evidence and mark inferences clearly, which prevents the agent from following malicious instructions potentially embedded in code comments.
[DATA_EXFILTRATION]: No network operations, API calls, or sensitive file access patterns were identified in the skill configuration or reference files.
[REMOTE_CODE_EXECUTION]: The skill is designed for documentation and analysis and does not include mechanisms for downloading or executing external scripts or packages.
[COMMAND_EXECUTION]: There are no shell commands or subprocess calls present in the skill, and it does not utilize the dynamic context injection syntax.

doc-generator