The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses uv run to execute various standard Python development tools including ruff, ty, mypy, pyright, rg, fd, and pytest. These tools are called using user-supplied file paths or modules as arguments.
[COMMAND_EXECUTION]: The skill executes a Python snippet via uv run python -c 'from dh_paths import plan_dir; print(plan_dir())' to resolve local paths. This involves importing and executing code from the local environment (e.g., the project being analyzed).
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) because it ingests and processes content from untrusted files provided by the user.
Ingestion points: Content from source code files, pyproject.toml, and project documentation.
Boundary markers: No explicit boundary markers or 'ignore' instructions are used when interpolating file content into prompts for analysis or planning.
Capability inventory: The skill has extensive capabilities including shell command execution (uv run), file writing for reports and plans, and the ability to delegate tasks to other sub-agents.
Sanitization: The skill does not perform sanitization or validation of the code content before it is analyzed by the AI to generate refactoring plans.

stinkysnake