The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process local documentation files which may contain untrusted data.
Ingestion points: The agent reads content from ai/memory-bank/requirements.md and ai/memory-bank/data-sources.md as part of its Step 1 workflow in SKILL.md.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are used when reading these files.
Capability inventory: The skill utilizes shell commands such as cat, ls, and grep to interact with the project file system.
Sanitization: There is no evidence of content validation or sanitization for the data ingested from these external files.

AI Engineer