git-worktree-clean

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes various git commands such as git worktree list, git worktree remove, git branch -d, and git push origin --delete to manage local development environments. It also utilizes standard shell utilities like du, sed, grep, and cut for disk usage reporting and text parsing.
  • [DATA_EXPOSURE]: Accesses local repository metadata, including branch names, commit history (via git log), and directory sizes (via du), to provide a summary of reclaimed space and cleanable resources.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from the local file system (branch names and worktree paths). While a maliciously crafted repository could theoretically include branch names with shell metacharacters, git's internal naming restrictions and the use of quotes around key variables significantly mitigate the risk of command injection.
  • Ingestion points: Reads output from git worktree list --porcelain and git rev-parse --abbrev-ref HEAD (SKILL.md).
  • Boundary markers: No explicit instructions to ignore embedded instructions are present.
  • Capability inventory: Performs file system writes (worktree removal), local branch deletion, and remote branch deletion (SKILL.md).
  • Sanitization: Most shell variables are enclosed in double quotes (e.g., "$WORKTREE", "$BRANCH"), though some loops rely on word splitting for list processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 11:27 AM