The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute local JavaScript utilities (check-file-size.mjs, install-git-hooks.mjs) using Node.js. These scripts are used to audit file sizes, identify splitting points, and install pre-commit hooks into the repository's configuration.
[EXTERNAL_DOWNLOADS]: Fetches the linguist-languages package from the npm registry. This package is a standard utility used to recognize source code languages and is required for the skill's file analysis logic.
[INDIRECT_PROMPT_INJECTION]: The skill involves the agent reading and analyzing the content of source files within a repository to determine where they should be split into smaller units based on "concepts."
Ingestion points: Untrusted source code files residing in the repository being analyzed.
Boundary markers: None specified in the instructions for the content of analyzed files.
Capability inventory: The skill can read/write files, modify git configuration/hooks, and install dependencies via package managers.
Sanitization: No sanitization or filtering of file content is described before the agent processes the code for structural analysis.

ratchet-files