The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted source code provided by users to generate agent instructions. This creates a surface for indirect prompt injection, where instructions hidden in the analyzed code could potentially influence the conversion process or the generated skill instructions.
Ingestion points: The skill reads source code provided by the user via file path or text input (SKILL.md).
Boundary markers: None specified in the instructions to separate untrusted code from analysis logic.
Capability inventory: The skill possesses file:read and file:write permissions to analyze input and save output.
Sanitization: No specific sanitization or filtering is mentioned for handling instructions embedded in the input code.
[COMMAND_EXECUTION]: The skill requires file:write permissions to save the generated SKILL.md files to the local file system. This capability is restricted to the skill's primary purpose of tool ingestion and skill creation.

ingest-tool