The Agent Skills Directory

[SAFE]: The skill performs legitimate documentation management tasks by analyzing codebase structure and metadata.
[DATA_EXFILTRATION]: The skill reads project configuration and documentation files to verify project structure and extract metadata. This access is required for its functionality and does not target sensitive system or credential files.
[COMMAND_EXECUTION]: The skill utilizes standard system commands to manage files and interact with the git repository. Evidence: Employs commands such as git, ls, cat, diff, cp, and ln for file operations and analysis.
[PROMPT_INJECTION]: The skill ingests untrusted data from the local codebase to generate documentation, creating an indirect prompt injection surface.
Ingestion points: Reads existing documentation and source files (e.g., README.md, src/ files) via cat and directory scanning.
Boundary markers: None explicitly implemented in the instructions.
Capability inventory: File system read/write access and network access for link validation via curl.
Sanitization: No explicit sanitization or filtering of external content is described.

md-docs