The Agent Skills Directory

[SAFE]: The skill performs legitimate documentation tasks by reading source code and writing documentation back to the project files. It does not attempt to access sensitive system files or environment variables.
[SAFE]: No network operations, external downloads, or data exfiltration patterns were detected. The skill operates entirely within the local filesystem context.
[COMMAND_EXECUTION]: The protocol includes a step to run local type-checking commands (e.g., typecheck) after modifying files to ensure that generated documentation (like JSDoc) does not introduce syntax errors. This is a standard development best practice and does not involve arbitrary or malicious command execution.
[PROMPT_INJECTION]: The instructions focus on guiding the AI's documentation style and quality criteria. There are no attempts to bypass safety filters, override system instructions, or extract internal prompts.

doc-gen