The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute git operations and manage directories. It provides script templates that interpolate user-provided variables (such as branch and worktree names) and repository metadata (such as remote URLs) into shell commands. If the AI agent does not correctly quote or sanitize these variables before execution, it could lead to command injection.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data which could lead to unexpected behavior if the data is maliciously crafted.
Ingestion points: User-supplied branch/worktree names and repository state retrieved via git remote get-url origin (file: SKILL.md).
Boundary markers: No explicit delimiters or instructions to treat data as untrusted are provided for the bash execution blocks.
Capability inventory: The skill uses Bash to perform directory creation (mkdir), complex string manipulation (sed, tr, printf), and various git commands (file: SKILL.md).
Sanitization: While the instructions include a directive to "Sanitize the directory name," there are no code-level sanitization routines or strict validation checks implemented within the provided script templates.

worktree