The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project data, such as source code and diff outputs, to generate documentation updates, which constitutes an indirect prompt injection surface.
Ingestion points: The agent is instructed to read source code, git diff output, and existing documentation files to perform its tasks.
Capability inventory: The agent has the capability to write to and update documentation files based on its reading of the project state.
Boundary markers: The instructions lack explicit markers or directions for the agent to distinguish between its instructions and potentially malicious content embedded in the files it reads.
Sanitization: No sanitization or validation process is specified for the data ingested from the project files.

documentation