The Agent Skills Directory

[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted data from the codebase being documented.
Ingestion points: The skill reads project files, source code, and existing documentation using Read, Glob, and Grep tools based on the target path provided in $ARGUMENTS.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when the agent interpolates content from project files into its context.
Capability inventory: The agent has access to Read, Write, Edit, Glob, Grep, and Bash tools, which are used to explore the repo and generate documentation files.
Sanitization: The VERIFICATION_CHECKLIST.md includes a mandatory step for the agent to verify that no secrets, tokens, or private endpoints are present in the final output.

document-writer