The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes source code files which serve as an untrusted ingestion point for external data.
Ingestion points: The skill ingests source code from various frameworks (Express.js, FastAPI, etc.) using the Read and Grep tools as defined in the frontmatter.
Boundary markers: The instructions do not specify any boundary markers or instructions for the agent to disregard natural language instructions that might be embedded within the source code comments or docstrings.
Capability inventory: The skill has the capability to Read existing files, Grep for patterns, and Write new documentation files to the filesystem.
Sanitization: There is no mention of sanitization or filtering logic to prevent malicious prompt injection strings within code comments from influencing the agent's behavior or the generated output.

api-documenter