The Agent Skills Directory

[SAFE]: The skill content is entirely instructional and provides safe templates for documentation. No malicious patterns or hardcoded credentials were identified.
[PROMPT_INJECTION]: The skill operates on external data, presenting an attack surface for indirect prompt injection.
Ingestion points: Uses read_file and search_with_context to process external documentation and source code.
Boundary markers: None defined to isolate external data from the agent's instructions.
Capability inventory: Includes write_file and apply_patch for modifying documents.
Sanitization: No validation or sanitization is performed on the content of the read files.

documentation-writing