The Agent Skills Directory

[Command Execution] (SAFE): The skill utilizes standard local Git commands (git worktree, git merge, git cherry-pick) and directory management (mkdir). These operations are consistent with the skill's primary purpose and do not involve privilege escalation or suspicious shell patterns.
[Indirect Prompt Injection] (LOW): The parallel execution workflow involves agents reading and summarizing code from different worktrees. Maliciously crafted code within the project could potentially influence the subagent's output. 1. Ingestion points: Local project files in trees/ directories. 2. Boundary markers: None provided in the example Task tool prompts. 3. Capability inventory: Git operations, file reading, and task delegation via the Task tool. 4. Sanitization: No explicit sanitization or instruction ignoring logic is present.

git-worktrees