The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes several standard development tools and package managers including git, npm, cargo, pip, poetry, and go to manage worktrees and dependencies. These commands are used as intended for creating isolated development environments.
[INDIRECT_PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection because it bases its logic on the presence and content of files that could be controlled by an external actor.
Ingestion points: The skill reads CLAUDE.md to find worktree directory preferences and detects the existence of package.json, Cargo.toml, requirements.txt, pyproject.toml, and go.mod to decide which setup commands to run.
Boundary markers: There are no explicit boundary markers or instructions to ignore potential commands embedded within these project files.
Capability inventory: The skill can execute shell commands for package installation, project building, and running tests (npm install, cargo test, pytest, etc.) as documented in the "Creation Steps" and "Quick Reference" sections of SKILL.md.
Sanitization: No specific sanitization or validation of the configuration file content is performed before triggering the associated command-line tools.

using-git-worktrees