The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs shell-based operations to manage Git worktrees and development environments. It uses git worktree add for isolation and triggers standard package managers such as npm, pip, poetry, cargo, and go upon detecting relevant manifest files. It also executes project test suites (npm test, pytest, etc.) to establish a clean baseline. These operations are the primary function of the skill.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by reading local project files. 1. Ingestion points: Reads CLAUDE.md for configuration and analyzes project manifests like package.json. 2. Boundary markers: No delimiters are used to isolate project data from instructions. 3. Capability inventory: The skill can execute various commands through package managers and test runners. 4. Sanitization: The skill does not sanitize content from the repository files before processing. This behavior is typical for development assistants operating on trusted local codebases.

using-git-worktrees