The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted documentation files which may contain instructions intended to bypass agent constraints through indirect prompt injection.\n- Ingestion points: The scripts/list_docs.py script reads any .md file found in the target directory (defaulting to docs/).\n- Boundary markers: The script outputs the parsed summaries without using delimiters or protective instructions to separate document content from agent instructions.\n- Capability inventory: The script is limited to file reading and standard output; it does not possess network access, file writing permissions, or the ability to execute subprocesses.\n- Sanitization: There is no sanitization or filtering of the content extracted from the YAML front-matter of the documentation files.

listing-docs