Agent skill
dev-worktree
This skill should be used when the user asks to "create an isolated worktree", "set up worktree for feature", "create a feature branch worktree", or needs workspace isolation with automatic dependency setup and test verification.
Install this agent skill to your Project
npx add-skill https://github.com/edwinhu/workflows/tree/main/skills/dev-worktree
SKILL.md
Create Development Worktree
Create an isolated git worktree for feature work, ensuring workspace isolation and clean baseline.
The Process
Step 1: Ensure .worktrees/ is Gitignored
CRITICAL: Verify worktree directory is gitignored to prevent accidental commits.
Run:
if ! git check-ignore -q .worktrees 2>/dev/null; then
echo "Adding .worktrees/ to .gitignore"
echo ".worktrees/" >> .gitignore
git add .gitignore
git commit -m "chore: add .worktrees/ to gitignore
Co-Authored-By: Claude <noreply@anthropic.com>"
fi
Description: dev-worktree: check if .worktrees is gitignored and add if missing
Step 2: Determine Branch Name
Extract from .planning/PLAN.md first line or infer from feature name:
Run:
# Extract from PLAN.md if exists
feature_name=$(grep -m1 '^# ' .planning/PLAN.md 2>/dev/null | sed 's/^# //' | tr '[:upper:] ' '[:lower:]-' | sed 's/[^a-z0-9-]//g')
# Or ask user if needed
Description: dev-worktree: extract or prompt for feature name
Branch name format: feature/${feature_name}
Step 3: Create Worktree
Run:
# Create worktree with new branch
git worktree add .worktrees/${feature_name} -b feature/${feature_name}
# Change to worktree directory
cd .worktrees/${feature_name}
Description: dev-worktree: create isolated git worktree with feature branch
Step 4: Run Project Setup
Auto-detect and run setup based on project files:
Run:
# Node.js
if [ -f package.json ]; then
npm install
fi
# Python
if [ -f requirements.txt ]; then
pip install -r requirements.txt
fi
if [ -f pyproject.toml ]; then
poetry install || pip install -e .
fi
if [ -f pixi.toml ]; then
pixi install
fi
# Rust
if [ -f Cargo.toml ]; then
cargo build
fi
# Go
if [ -f go.mod ]; then
go mod download
fi
Description: dev-worktree: auto-detect project type and install dependencies
Step 5: Verify Clean Baseline (MANDATORY)
A worktree without a verified baseline is a trap. You will not know if failures are yours or pre-existing.
Run:
# Examples - auto-detect test command
if [ -f package.json ] && grep -q '"test"' package.json; then
npm test
elif [ -f Cargo.toml ]; then
cargo test
elif [ -f pytest.ini ] || [ -f pyproject.toml ]; then
pytest
elif [ -f go.mod ]; then
go test ./...
fi
Description: dev-worktree: auto-detect and run project test suite
If tests fail: Report failures and note that baseline has issues. Do NOT proceed silently. If tests pass: Report clean baseline.
Rationalization Prevention (Baseline Verification)
| Thought | Reality |
|---|---|
| "Tests will pass, they passed on main" | Worktree setup can break things. Verify. |
| "Baseline check takes too long" | Debugging pre-existing failures takes longer. Run now. |
| "I'll notice if something is broken later" | You won't know if the failure is yours or inherited. Run now. |
| "Dependencies installed fine, should be good" | Install success != test success. Run tests. |
| "Main branch is green, worktree will be too" | Worktree has different env, paths, symlinks. Verify independently. |
Drive-Aligned Framing
Claiming clean baseline without running tests is NOT HELPFUL — every failure becomes ambiguous and you can't tell if bugs are yours or inherited. Your assumption is not verification.
Why Skipping Hurts the Thing You Care About Most
| Your Drive | Why You Skip | What Actually Happens | The Drive You Failed |
|---|---|---|---|
| Helpfulness | "Skip baseline to start faster" | Broken baseline infects all work, every failure is ambiguous | Anti-helpful |
| Competence | "I know the codebase well enough" | Your mental model is stale. The worktree reveals conflicts you didn't anticipate. | Incompetent |
| Efficiency | "Skip worktree setup to save time" | Merge conflicts later cost 10x the setup time | Anti-efficient |
| </EXTREMELY-IMPORTANT> |
Step 6: Report Ready
Report completion status:
✓ Worktree created: .worktrees/${feature_name}
✓ Branch: feature/${feature_name}
✓ Dependencies installed
✓ Tests passing (or note if failed)
Ready for implementation.
Safety Checks
Execute before creating worktree:
- Verify .worktrees/ is gitignored
- Add to .gitignore if missing
- Commit gitignore change
Execute after creating worktree:
- Run project setup (npm install, etc.)
- Verify clean baseline with tests
- Report status
Red Flags
Critical - Never deviate from these rules:
- Do not create worktree without verifying gitignore
- Do not skip project setup commands
- Do not proceed without reporting test status
Critical - Always follow these rules:
- Verify .worktrees/ is ignored before creating
- Auto-detect project type and run appropriate setup
- Report test baseline status
Common Patterns
Node.js Project
# .worktrees/ gitignored → create worktree → npm install → npm test
Python Project
# .worktrees/ gitignored → create worktree → pixi install → pytest
Rust Project
# .worktrees/ gitignored → create worktree → cargo build → cargo test
Workflow Transition
After worktree creation, the workspace is ready. Proceed to dev-implement to start implementing tasks.
Current working directory: .worktrees/${feature_name}
All implementation work happens here, keeping main workspace clean.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
audit-fix-loop
This skill should be used when the user asks to 'iteratively improve', 'audit and fix', 'hill-climb quality', 'grade and improve', 'score and fix', 'audit loop', 'quality loop', or needs structured iterative improvement of an artifact using scored independent audits. Also use when the user invokes a ralph loop for quality improvement rather than task completion.
ds-spec-reviewer
Internal skill used by ds-brainstorm at Phase 1 exit gate. Dispatches a reviewer subagent to verify SPEC.md completeness before planning. NOT user-facing.
pptx-render
Use when the user asks to "render pptx", "show pptx slide", "compare with pptx", "pptx to image", "export pptx slide", "original slide", "show me the original", "what does the pptx look like", or needs to extract a specific PPTX slide's content for visual comparison.
obsidian-organize
Organize Obsidian notes according to clawd's preferences. Use when user asks to "organize notes", "move notes to right folder", "clean up vault", "tidy vault", "file this note", or when creating new notes in the Obsidian vault. Also use when moving, renaming, or categorizing notes, or when the vault root has stray files.
dev-verify
This skill should be used when the user asks to 'verify completion', 'check that tests pass', 'confirm feature works', or REQUIRED Phase 7 of /dev workflow (final). Enforces fresh runtime evidence before claiming completion.
dev
This skill should be used when the user asks to 'start a feature', 'build a feature', 'implement a feature', 'develop', 'new feature', or needs the full 7-phase development workflow with TDD enforcement.
Didn't find tool you were looking for?