The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several shell commands to manage the git environment and automate workflows. These include git worktree add, directory manipulation, and committing changes to .gitignore when necessary to prevent accidental tracking of worktree contents.
[EXTERNAL_DOWNLOADS]: The skill automatically triggers external downloads from official package registries by invoking standard package managers such as npm install, pip install, poetry install, cargo build, and go mod download. These actions occur only when the corresponding project configuration files (e.g., package.json, requirements.txt) are present in the repository.
[PROMPT_INJECTION]: The skill implements logic to read configuration preferences from a CLAUDE.md file using grep. This represents an indirect prompt injection surface where a project file can influence the agent's directory selection logic.
Ingestion points: Reads directory preferences from CLAUDE.md via shell grep.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands in the ingested data are present.
Capability inventory: The skill has the capability to execute commands (npm, pip, cargo, go) and modify the git repository (git commit).
Sanitization: No sanitization or validation of the content extracted from CLAUDE.md is performed before use.

using-git-worktrees