The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes local Python scripts (scripts/scan.py and scripts/check_doc.py) that execute shell commands to retrieve repository information. Specifically, it uses subprocess.run to call git for listing files, retrieving logs, and identifying high-churn files. These operations are scoped to project discovery and documentation validation.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the user's codebase, which could theoretically contain malicious instructions.
Ingestion points: Reads repository manifest files (package.json, go.mod, etc.), source code, and existing documentation files during scanning and checking phases.
Boundary markers: The instructions do not explicitly mandate delimiters or 'ignore' instructions for the processed data, relying on the agent's internal reasoning.
Capability inventory: The skill executes shell commands via local scripts (git queries) and can write files to the target repository's documentation directories.
Sanitization: scripts/check_doc.py contains a robust credential scanner that checks for API keys, tokens, and private keys in the generated documentation to prevent accidental data exposure.

documentation-writer