2.1k
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill max-sixty/worktrunk --skill worktrunk- SKILL.md9.0 KB
Overview
This skill provides practical guidance for Worktrunk, a CLI for managing Git worktrees and automating developer workflows. It focuses on configuring user and project settings, setting up LLM-based commit message generation, designing hooks, and troubleshooting common issues. Use it to configure your environment or to add team-wide automation safely and reproducibly.
How this skill works
I explain the two config scopes: user config (~/.config/worktrunk/config.toml) for personal preferences and LLM integrations, and project config (<repo>/.config/wt.toml) for team-shared hooks and automation. For each task I show how to detect tools, validate commands, propose or apply changes, and recommend tests to verify behavior. I also cover permission models, safe defaults, and patterns for spawning agent handoffs in supported multiplexers.
When to use it
- Set up LLM commit message generation for your local environment
- Add project hooks that run on worktree create, switch, pre-commit, or pre-merge
- Automate dependency install or test runs for all contributors
- Troubleshoot hook failures, LLM integration errors, or shell-integration problems
- Spawn a background agent session in tmux or Zellij for parallel work (explicit request)
Best practices
- Always validate commands exist (which npm, cargo, etc.) before adding hooks
- For user config, propose changes and get explicit consent before editing
- For project config, add explanatory comments and prefer reversible, non-destructive steps
- Avoid dangerous commands in hooks (rm -rf, DROP TABLE, sudo, uncontrolled curl)
- Test new hooks by creating a worktree (wt switch --create test-hooks) and by running dry-runs
Example use cases
- Enable LLM commit generation by detecting available tools and adding a commit.generation command to ~/.config/worktrunk/config.toml
- Add a post-create hook that runs npm install when a new worktree is created
- Add a pre-merge hook to run the test suite and block merges on failure
- Convert a single-command hook into a named table to add extra steps (install + migrate)
- Spawn a tmux session that creates a worktree and hands it to Claude for background agent work
FAQ
Commit message generation is a personal preference—add it to your user config at ~/.config/worktrunk/config.toml after validating the LLM tool is installed.
Can project hooks be edited directly?
Yes. Project config (.config/wt.toml) is designed to be committed; changes are proactive but should validate commands and include comments explaining the automation.
How do I test a new hook safely?
Validate the command exists, then create a test worktree with wt switch --create test-hooks and observe hook behavior; keep destructive actions out of tests.