The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (scripts/codebase-map.py and scripts/find-god-files.py) provided within the plugin's own script directory to perform codebase mapping and detect large files. These operations are local to the environment and serve established analysis functions.\n- [SAFE]: The skill identifies and manages an indirect prompt injection surface by ingesting external data from project source files, audit reports, and RFCs. It includes mandatory safety rules from references/subagent-safety.md which instruct the agent to treat repository content as data rather than instructions. Its capabilities are restricted to local script execution and writing plan documentation to the docs/arc/plans/ directory using structured XML templates.

improve