daft

Appearance

Agent Skill ​

daft provides an Agent Skill that teaches AI coding agents the daft worktree workflow. The skill follows an open standard and works with any agent that supports skills, including Claude Code, Cursor, Windsurf, and others.

What Is It ​

The daft-worktree-workflow skill is a SKILL.md file that agents load as context when working in daft-managed repositories. It contains structured knowledge about:

  • The worktree-centric development philosophy
  • All daft commands and when to use each one
  • The daft.yml hooks system for automating worktree setup
  • Environment tool detection (mise, direnv, nvm, pyenv, and more)
  • How to suggest automation for projects that lack it
  • Correct worktree-aware translations for common Git operations

When loaded, the agent understands that "create a branch" means daft start, that each worktree needs its own dependency install, and that daft.yml hooks can automate the setup process.

Installation ​

bash
npx skills add avihut/daft

This clones the skill from the daft repository and installs it into your agent's skills directory.

Manual -- User-Global ​

Copy the SKILL.md file to your agent's skills directory. For example, with Claude Code:

bash
mkdir -p ~/.claude/skills/daft-worktree-workflow
curl -o ~/.claude/skills/daft-worktree-workflow/SKILL.md \
  https://raw.githubusercontent.com/avihut/daft/master/SKILL.md

The skill will be available in all sessions. Consult your agent's documentation for the correct skills directory if using a different tool.

Manual -- Project-Level ​

To include the skill for a specific project, copy it into the project's skills directory:

bash
mkdir -p .claude/skills/daft-worktree-workflow
cp /path/to/SKILL.md .claude/skills/daft-worktree-workflow/SKILL.md

Commit it to the repository so all contributors benefit.

When the Skill Activates ​

The skill activates automatically when the agent finds it in its skills search path. This happens when:

  • The skill is installed via npx skills add or manually placed in the skills directory
  • The user is working in a daft-managed repository (bare .git/ with worktree siblings)
  • The user asks about worktree workflows, daft commands, or environment isolation

The skill can also be invoked explicitly by the user.

What the Skill Teaches ​

Detecting daft Repositories ​

The agent learns to recognize the daft directory layout: a bare .git/ directory at the project root with branch worktrees as sibling directories.

Command Translation ​

Instead of suggesting git checkout -b, the agent suggests daft start. Instead of git switch, it suggests navigating to the worktree directory. The skill maps common Git intents to their daft equivalents.

Hooks Automation ​

The skill covers the full daft.yml configuration format: hook types, execution modes (parallel, piped, follow), job definitions, dependencies, template variables, skip/only conditions, and trust management.

Environment Tooling ​

When the agent encounters a daft repo, it checks for environment tool markers (.mise.toml, .envrc, .nvmrc, package.json, Cargo.toml, etc.) and suggests daft.yml hooks that automate tool setup for new worktrees.

Per-Worktree Isolation ​

The skill emphasizes that each worktree is a fully isolated workspace. Dependencies, build artifacts, and environment config are not shared. This means npm install must run in each worktree, virtual environments must be created separately, and so on.

Manual Integration ​

If you prefer not to use the skills system, you can reference the skill content directly in your agent's project instructions file (e.g., CLAUDE.md for Claude Code). The SKILL.md file at the repository root is the source of truth. Copy relevant sections into your project's instructions file.

See Also ​

Released under the MIT License.

Copyright © 2025-present Avihu Turzion