Worktree Manager

Global Worktree Manager

Manage parallel development across ALL projects using git worktrees with Claude Code agents. Each worktree is an isolated copy of the repo on a different branch, stored centrally at ~/tmp/worktrees/.

IMPORTANT: You (Claude) can perform ALL operations manually using standard tools (jq, git, bash). Scripts are helpers, not requirements. If a script fails, fall back to manual operations described in this document.

When This Skill Activates

Trigger phrases:

• "spin up worktrees for X, Y, Z"
• "create 3 worktrees for features A, B, C"
• "new worktree for feature/auth"
• "what's the status of my worktrees?"
• "show all worktrees" / "show worktrees for this project"
• "clean up merged worktrees"
• "clean up the auth worktree"
• "launch agent in worktree X"
• "create worktree with sparse-checkout for apps/api"
• "spin up a sparse worktree for the monorepo"
• "add apps/frontend to the sparse-checkout"
• "disable sparse-checkout on the auth worktree"
• "show sparse-checkout dirs for my worktrees"

File Locations

Core Concepts

Centralized Worktree Storage

All worktrees live in ~/tmp/worktrees/<project-name>/<branch-slug>/

~/tmp/worktrees/
├── obsidian-ai-agent/
│   ├── feature-auth/           # branch: feature/auth
│   ├── feature-payments/       # branch: feature/payments
│   └── fix-login-bug/          # branch: fix/login-bug
└── another-project/
    └── feature-dark-mode/

Branch Slug Convention

Branch names are slugified for filesystem safety by replacing / with -:

• feature/auth → feature-auth
• fix/login-bug → fix-login-bug
• feat/user-profile → feat-user-profile

Slugify manually: echo "feature/auth" | tr '/' '-' → feature-auth

Sparse-Checkout Support

For monorepos, sparse-checkout lets you check out only the directories you need — saving disk space and improving IDE performance.

How it works: After creating a worktree with --no-checkout, run git sparse-checkout init --cone then git sparse-checkout set <dirs> to populate only the specified directories.

Config priority (highest → lowest):

1. Explicit directories passed during creation
2. Project config (.claude/worktree.json → sparseCheckout.directories)
3. Skill config (config.json → sparseCheckout.defaultDirectories)
4. Disabled (full checkout) if nothing configured

Port Allocation Rules

• Global pool: 8100-8199 (100 ports total)
• Per worktree: 2 ports allocated (for API + frontend patterns)
• Globally unique: Ports are tracked globally to avoid conflicts across projects
• Check before use: Always verify port isn't in use by system: lsof -i :<port>

Global Registry

Location

~/.claude/worktree-registry.json

Schema

{
  "worktrees": [
    {
      "id": "unique-uuid",
      "project": "obsidian-ai-agent",
      "repoPath": "/Users/rasmus/Projects/obsidian-ai-agent",
      "branch": "feature/auth",
      "branchSlug": "feature-auth",
      "worktreePath": "/Users/rasmus/tmp/worktrees/obsidian-ai-agent/feature-auth",
      "ports": [8100, 8101],
      "createdAt": "2025-12-04T10:00:00Z",
      "validatedAt": "2025-12-04T10:02:00Z",
      "agentLaunchedAt": "2025-12-04T10:03:00Z",
      "task": "Implement OAuth login",
      "prNumber": null,
      "status": "active",
      "sparseCheckout": {
        "enabled": true,
        "directories": ["apps/api", "packages/core"]
      }
    }
  ],
  "portPool": {
    "start": 8100,
    "end": 8199,
    "allocated": [8100, 8101]
  }
}

Field Descriptions

Worktree entry fields:

Port pool fields:

Manual Registry Operations

Read entire registry:

cat ~/.claude/worktree-registry.json | jq '.'

List all worktrees:

cat ~/.claude/worktree-registry.json | jq '.worktrees[]'

List worktrees for specific project:

cat ~/.claude/worktree-registry.json | jq '.worktrees[] | select(.project == "my-project")'

Get allocated ports:

cat ~/.claude/worktree-registry.json | jq '.portPool.allocated'

Find worktree by branch (partial match):

cat ~/.claude/worktree-registry.json | jq '.worktrees[] | select(.branch | contains("auth"))'

Add worktree entry manually:

TMP=$(mktemp)
jq '.worktrees += [{
  "id": "'$(uuidgen)'",
  "project": "my-project",
  "repoPath": "/path/to/repo",
  "branch": "feature/auth",
  "branchSlug": "feature-auth",
  "worktreePath": "/Users/me/tmp/worktrees/my-project/feature-auth",
  "ports": [8100, 8101],
  "createdAt": "'$(date -u +%Y-%m-%dT%H:%M:%SZ)'",
  "validatedAt": null,
  "agentLaunchedAt": null,
  "task": "My task",
  "prNumber": null,
  "status": "active",
  "sparseCheckout": {"enabled": false, "directories": []}
}]' ~/.claude/worktree-registry.json > "$TMP" && mv "$TMP" ~/.claude/worktree-registry.json

Add ports to allocated pool:

TMP=$(mktemp)
jq '.portPool.allocated += [8100, 8101] | .portPool.allocated |= unique | .portPool.allocated |= sort_by(.)' \
  ~/.claude/worktree-registry.json > "$TMP" && mv "$TMP" ~/.claude/worktree-registry.json

Remove worktree entry:

TMP=$(mktemp)
jq 'del(.worktrees[] | select(.project == "my-project" and .branch == "feature/auth"))' \
  ~/.claude/worktree-registry.json > "$TMP" && mv "$TMP" ~/.claude/worktree-registry.json

Release ports from pool:

TMP=$(mktemp)
jq '.portPool.allocated = (.portPool.allocated | map(select(. != 8100 and . != 8101)))' \
  ~/.claude/worktree-registry.json > "$TMP" && mv "$TMP" ~/.claude/worktree-registry.json

Initialize empty registry (if missing):

mkdir -p ~/.claude
cat > ~/.claude/worktree-registry.json << 'EOF'
{
  "worktrees": [],
  "portPool": {
    "start": 8100,
    "end": 8199,
    "allocated": []
  }
}
EOF

Manual Port Allocation

If scripts/allocate-ports.sh fails, allocate ports manually:

Step 1: Get currently allocated ports

ALLOCATED=$(cat ~/.claude/worktree-registry.json | jq -r '.portPool.allocated[]' | sort -n)
echo "Currently allocated: $ALLOCATED"

Step 2: Find first available port (not in allocated list AND not in use by system)

for PORT in $(seq 8100 8199); do
  # Check if in registry
  if ! echo "$ALLOCATED" | grep -q "^${PORT}$"; then
    # Check if in use by system
    if ! lsof -i :"$PORT" &>/dev/null; then
      echo "Available: $PORT"
      break
    fi
  fi
done

Step 3: Add to allocated pool

TMP=$(mktemp)
jq '.portPool.allocated += [8100] | .portPool.allocated |= unique | .portPool.allocated |= sort_by(.)' \
  ~/.claude/worktree-registry.json > "$TMP" && mv "$TMP" ~/.claude/worktree-registry.json

What You (Claude) Do vs What Scripts Do

Workflows

1. Create Multiple Worktrees with Agents

User says: "Spin up 3 worktrees for feature/auth, feature/payments, and fix/login-bug"

You do (can parallelize with subagents):

For EACH branch (can run in parallel):

1. SETUP
   a. Get project name:
      PROJECT=$(basename $(git remote get-url origin 2>/dev/null | sed 's/\.git$//') 2>/dev/null || basename $(pwd))
   b. Get repo root:
      REPO_ROOT=$(git rev-parse --show-toplevel)
   c. Slugify branch:
      BRANCH_SLUG=$(echo "feature/auth" | tr '/' '-')
   d. Determine worktree path:
      WORKTREE_PATH=~/tmp/worktrees/$PROJECT/$BRANCH_SLUG

2. ALLOCATE PORTS
   Option A (script): ~/.claude/skills/worktree-manager/scripts/allocate-ports.sh 2
   Option B (manual): Find 2 unused ports from 8100-8199, add to registry

2.5. DETERMINE SPARSE-CHECKOUT (if monorepo)
   a. Check if user specified directories explicitly
   b. Check .claude/worktree.json for sparseCheckout.directories
   c. Check config.json for sparseCheckout.defaultDirectories
   d. If any found and sparseCheckout enabled → use --no-checkout in step 3

3. CREATE WORKTREE
   # Without sparse-checkout:
   mkdir -p ~/tmp/worktrees/$PROJECT
   git worktree add $WORKTREE_PATH -b $BRANCH

   # With sparse-checkout:
   mkdir -p ~/tmp/worktrees/$PROJECT
   git worktree add $WORKTREE_PATH -b $BRANCH --no-checkout
   ~/.claude/skills/worktree-manager/scripts/sparse-checkout.sh setup $WORKTREE_PATH <dir1> <dir2>...
   # Or manually:
   # git -C $WORKTREE_PATH sparse-checkout init --cone
   # git -C $WORKTREE_PATH sparse-checkout set <dir1> <dir2>...

   # If branch exists already, omit -b flag

4. COPY UNCOMMITTED RESOURCES
   cp -r .agents $WORKTREE_PATH/ 2>/dev/null || true
   cp .env.example $WORKTREE_PATH/.env 2>/dev/null || true

5. INSTALL DEPENDENCIES
   cd $WORKTREE_PATH
   # Detect and run: npm install / uv sync / etc.

6. VALIDATE (start server, health check, stop)
   a. Start server with allocated port
   b. Wait and health check: curl -sf http://localhost:$PORT/health
   c. Stop server
   d. If FAILS: report error but continue with other worktrees

7. REGISTER IN GLOBAL REGISTRY
   Option A (script): ~/.claude/skills/worktree-manager/scripts/register.sh ...
   Option B (manual): Update ~/.claude/worktree-registry.json with jq

8. LAUNCH AGENT
   Option A (script): ~/.claude/skills/worktree-manager/scripts/launch-agent.sh $WORKTREE_PATH "task"
   Option B (manual): Open terminal manually, cd to path, run claude

AFTER ALL COMPLETE:
- Report summary table to user
- Note any failures with details

2. Check Status

With script:

~/.claude/skills/worktree-manager/scripts/status.sh
~/.claude/skills/worktree-manager/scripts/status.sh --project my-project

Manual:

# All worktrees
cat ~/.claude/worktree-registry.json | jq -r '.worktrees[] | "\(.project)\t\(.branch)\t\(.ports | join(","))\t\(.status)\t\(.task // "-")"'

# For current project
PROJECT=$(basename $(git remote get-url origin 2>/dev/null | sed 's/\.git$//'))
cat ~/.claude/worktree-registry.json | jq -r ".worktrees[] | select(.project == \"$PROJECT\") | \"\(.branch)\t\(.ports | join(\",\"))\t\(.status)\""

3. Launch Agent Manually

If launch-agent.sh fails:

For Ghostty:

open -na "Ghostty.app" --args -e bash -c "cd '$WORKTREE_PATH' && claude --dangerously-skip-permissions"

For iTerm2:

osascript -e 'tell application "iTerm2" to create window with default profile' \
  -e 'tell application "iTerm2" to tell current session of current window to write text "cd '"$WORKTREE_PATH"' && claude"'

For tmux:

tmux new-session -d -s "wt-$PROJECT-$BRANCH_SLUG" -c "$WORKTREE_PATH" "bash -c 'claude --dangerously-skip-permissions'"

4. Cleanup Worktree

With script:

~/.claude/skills/worktree-manager/scripts/cleanup.sh my-project feature/auth --delete-branch

Manual cleanup:

# 1. Get worktree info from registry
ENTRY=$(cat ~/.claude/worktree-registry.json | jq '.worktrees[] | select(.project == "my-project" and .branch == "feature/auth")')
WORKTREE_PATH=$(echo "$ENTRY" | jq -r '.worktreePath')
PORTS=$(echo "$ENTRY" | jq -r '.ports[]')
REPO_PATH=$(echo "$ENTRY" | jq -r '.repoPath')

# 2. Kill processes on ports
for PORT in $PORTS; do
  lsof -ti:"$PORT" | xargs kill -9 2>/dev/null || true
done

# 3. Remove worktree
cd "$REPO_PATH"
git worktree remove "$WORKTREE_PATH" --force 2>/dev/null || rm -rf "$WORKTREE_PATH"
git worktree prune

# 4. Remove from registry
TMP=$(mktemp)
jq 'del(.worktrees[] | select(.project == "my-project" and .branch == "feature/auth"))' \
  ~/.claude/worktree-registry.json > "$TMP" && mv "$TMP" ~/.claude/worktree-registry.json

# 5. Release ports
TMP=$(mktemp)
for PORT in $PORTS; do
  jq ".portPool.allocated = (.portPool.allocated | map(select(. != $PORT)))" \
    ~/.claude/worktree-registry.json > "$TMP" && mv "$TMP" ~/.claude/worktree-registry.json
done

# 6. Optionally delete branch
git branch -D feature/auth
git push origin --delete feature/auth

Manual Sparse-Checkout Setup

If scripts/sparse-checkout.sh fails, set up sparse-checkout manually:

Initialize sparse-checkout in an existing worktree:

cd $WORKTREE_PATH
git sparse-checkout init --cone
git sparse-checkout set apps/api packages/core packages/shared

Add more directories later:

cd $WORKTREE_PATH
# Get current directories
CURRENT=$(git sparse-checkout list)
# Set all directories (current + new)
git sparse-checkout set $CURRENT apps/frontend

View current sparse-checkout directories:

cd $WORKTREE_PATH
git sparse-checkout list

Disable sparse-checkout (restore full checkout):

cd $WORKTREE_PATH
git sparse-checkout disable

Create worktree with sparse-checkout from scratch:

# 1. Create worktree without checking out files
git worktree add ~/tmp/worktrees/$PROJECT/$BRANCH_SLUG -b $BRANCH --no-checkout

# 2. Initialize sparse-checkout
git -C ~/tmp/worktrees/$PROJECT/$BRANCH_SLUG sparse-checkout init --cone

# 3. Set directories to check out
git -C ~/tmp/worktrees/$PROJECT/$BRANCH_SLUG sparse-checkout set apps/api packages/core

# 4. Files are now populated only in specified directories

Package Manager Detection

Detect by checking for lockfiles in priority order:

Detection logic:

cd $WORKTREE_PATH
if [ -f "bun.lockb" ]; then bun install
elif [ -f "pnpm-lock.yaml" ]; then pnpm install
elif [ -f "yarn.lock" ]; then yarn install
elif [ -f "package-lock.json" ]; then npm install
elif [ -f "uv.lock" ]; then uv sync
elif [ -f "pyproject.toml" ]; then uv sync
elif [ -f "requirements.txt" ]; then pip install -r requirements.txt
elif [ -f "go.mod" ]; then go mod download
elif [ -f "Cargo.toml" ]; then cargo build
fi

Dev Server Detection

Look for dev commands in this order:

1. docker-compose.yml / compose.yml: docker-compose up -d or docker compose up -d
2. package.json scripts: Look for dev, start:dev, serve
3. Python with uvicorn: uv run uvicorn app.main:app --port $PORT
4. Python with Flask: flask run --port $PORT
5. Go: go run .

Port injection: Most servers accept PORT env var or --port flag

Project-Specific Config (Optional)

Projects can provide .claude/worktree.json for custom settings:

{
  "ports": {
    "count": 2,
    "services": ["api", "frontend"]
  },
  "install": "uv sync && cd frontend && npm install",
  "validate": {
    "start": "docker-compose up -d",
    "healthCheck": "curl -sf http://localhost:{{PORT}}/health",
    "stop": "docker-compose down"
  },
  "copyDirs": [".agents", ".env.example", "data/fixtures"],
  "sparseCheckout": {
    "enabled": true,
    "directories": ["apps/api", "apps/frontend", "packages/core", "packages/shared"]
  }
}

If this file exists, use its settings. Otherwise, auto-detect. When sparseCheckout.enabled is true, worktrees will be created with --no-checkout and only the listed directories will be checked out.

Parallel Worktree Creation

When creating multiple worktrees, use subagents for parallelization:

User: "Spin up worktrees for feature/a, feature/b, feature/c"

You:
1. Allocate ports for ALL worktrees upfront (6 ports total)
2. Spawn 3 subagents, one per worktree
3. Each subagent:
   - Creates its worktree
   - Installs deps
   - Validates
   - Registers (with its pre-allocated ports)
   - Launches agent
4. Collect results from all subagents
5. Report unified summary with any failures noted

Safety Guidelines

1. Before cleanup, check PR status:

   • PR merged → safe to clean everything
   • PR open → warn user, confirm before proceeding
   • No PR → warn about unsubmitted work

2. Before deleting branches, confirm if:

   • PR not merged
   • No PR exists
   • Worktree has uncommitted changes

3. Port conflicts: If port in use by non-worktree process, pick different port

4. Orphaned worktrees: If original repo deleted, mark as orphaned in status

5. Max worktrees: With 100-port pool and 2 ports each, max ~50 concurrent worktrees

Script Reference

Scripts are in ~/.claude/skills/worktree-manager/scripts/

allocate-ports.sh

~/.claude/skills/worktree-manager/scripts/allocate-ports.sh <count>
# Returns: space-separated port numbers (e.g., "8100 8101")
# Automatically updates registry

sparse-checkout.sh

~/.claude/skills/worktree-manager/scripts/sparse-checkout.sh setup <worktree-path> <dir1> [dir2]...
~/.claude/skills/worktree-manager/scripts/sparse-checkout.sh list <worktree-path>
~/.claude/skills/worktree-manager/scripts/sparse-checkout.sh add <worktree-path> <dir1> [dir2]...
~/.claude/skills/worktree-manager/scripts/sparse-checkout.sh disable <worktree-path>
# Example:
~/.claude/skills/worktree-manager/scripts/sparse-checkout.sh setup \
  "$HOME/tmp/worktrees/my-project/feature-auth" apps/api packages/core

register.sh

~/.claude/skills/worktree-manager/scripts/register.sh \
  <project> <branch> <branch-slug> <worktree-path> <repo-path> <ports> [task] [sparse-enabled] [sparse-dirs]
# Example:
~/.claude/skills/worktree-manager/scripts/register.sh \
  "my-project" "feature/auth" "feature-auth" \
  "$HOME/tmp/worktrees/my-project/feature-auth" \
  "/path/to/repo" "8100,8101" "Implement OAuth"
# Example with sparse-checkout:
~/.claude/skills/worktree-manager/scripts/register.sh \
  "my-project" "feature/auth" "feature-auth" \
  "$HOME/tmp/worktrees/my-project/feature-auth" \
  "/path/to/repo" "8100,8101" "Implement OAuth" true "apps/api,packages/core"

launch-agent.sh

~/.claude/skills/worktree-manager/scripts/launch-agent.sh <worktree-path> [task]
# Opens new terminal window (Ghostty by default) with Claude Code

status.sh

~/.claude/skills/worktree-manager/scripts/status.sh [--project <name>]
# Shows all worktrees, or filtered by project

cleanup.sh

~/.claude/skills/worktree-manager/scripts/cleanup.sh <project> <branch> [--delete-branch]
# Kills ports, removes worktree, updates registry
# --delete-branch also removes local and remote git branches

release-ports.sh

~/.claude/skills/worktree-manager/scripts/release-ports.sh <port1> [port2] ...
# Releases ports back to pool

Skill Config

Location: ~/.claude/skills/worktree-manager/config.json

{
  "terminal": "ghostty",
  "shell": "bash",
  "claudeCommand": "claude --dangerously-skip-permissions",
  "portPool": {
    "start": 8100,
    "end": 8199
  },
  "portsPerWorktree": 2,
  "worktreeBase": "~/tmp/worktrees",
  "defaultCopyDirs": [".agents", ".env.example"],
  "sparseCheckout": {
    "enabled": false,
    "defaultDirectories": []
  }
}

Options:

• terminal: ghostty, iterm2, tmux, wezterm, kitty, alacritty
• shell: bash, zsh, fish (adjust syntax in claudeCommand if using fish)
• claudeCommand: The command to launch Claude Code (default uses --dangerously-skip-permissions for autonomous operation)

Common Issues

"Worktree already exists"

git worktree list
git worktree remove <path> --force
git worktree prune

"Branch already exists"

# Use existing branch (omit -b flag)
git worktree add <path> <branch>

"Port already in use"

lsof -i :<port>
# Kill if stale, or pick different port

Registry out of sync

# Compare registry to actual worktrees
cat ~/.claude/worktree-registry.json | jq '.worktrees[].worktreePath'
find ~/tmp/worktrees -maxdepth 2 -type d

# Remove orphaned entries or add missing ones

Validation failed

1. Check stderr/logs for error message
2. Common issues: missing env vars, database not running, wrong port
3. Report to user with details
4. Continue with other worktrees
5. User can fix and re-validate manually

Sparse-checkout shows unexpected files

# Re-initialize cone mode
cd $WORKTREE_PATH
git sparse-checkout init --cone
git sparse-checkout set <dir1> <dir2>

Sparse-checkout disabled accidentally

# Re-enable with the same directories from registry
DIRS=$(cat ~/.claude/worktree-registry.json | jq -r '.worktrees[] | select(.worktreePath == "'$WORKTREE_PATH'") | .sparseCheckout.directories[]')
cd $WORKTREE_PATH
git sparse-checkout init --cone
git sparse-checkout set $DIRS

Need a directory not in sparse-checkout

# Add it without disrupting existing checkout
~/.claude/skills/worktree-manager/scripts/sparse-checkout.sh add $WORKTREE_PATH apps/new-dir
# Or manually:
cd $WORKTREE_PATH
CURRENT=$(git sparse-checkout list)
git sparse-checkout set $CURRENT apps/new-dir

Example Session

User: "Spin up 2 worktrees for feature/dark-mode and fix/login-bug"

You:

1. Detect project: obsidian-ai-agent (from git remote)
2. Detect package manager: uv (found uv.lock)
3. Allocate 4 ports: ~/.claude/skills/worktree-manager/scripts/allocate-ports.sh 4 → 8100 8101 8102 8103
4. Create worktrees:

   mkdir -p ~/tmp/worktrees/obsidian-ai-agent
   git worktree add ~/tmp/worktrees/obsidian-ai-agent/feature-dark-mode -b feature/dark-mode
   git worktree add ~/tmp/worktrees/obsidian-ai-agent/fix-login-bug -b fix/login-bug
   

5. Copy .agents/:

   cp -r .agents ~/tmp/worktrees/obsidian-ai-agent/feature-dark-mode/
   cp -r .agents ~/tmp/worktrees/obsidian-ai-agent/fix-login-bug/
   

6. Install deps in each worktree:

   (cd ~/tmp/worktrees/obsidian-ai-agent/feature-dark-mode && uv sync)
   (cd ~/tmp/worktrees/obsidian-ai-agent/fix-login-bug && uv sync)
   

7. Validate each (start server, health check, stop)
8. Register both worktrees in ~/.claude/worktree-registry.json
9. Launch agents:

   ~/.claude/skills/worktree-manager/scripts/launch-agent.sh \
     ~/tmp/worktrees/obsidian-ai-agent/feature-dark-mode "Implement dark mode toggle"
   ~/.claude/skills/worktree-manager/scripts/launch-agent.sh \
     ~/tmp/worktrees/obsidian-ai-agent/fix-login-bug "Fix login redirect bug"
   

10. Report:

    Created 2 worktrees with agents:
    
    | Branch | Ports | Path | Task |
    |--------|-------|------|------|
    | feature/dark-mode | 8100, 8101 | ~/tmp/worktrees/.../feature-dark-mode | Implement dark mode |
    | fix/login-bug | 8102, 8103 | ~/tmp/worktrees/.../fix-login-bug | Fix login redirect |
    
    Both agents running in Ghostty windows.
    

Monorepo with Sparse-Checkout

User: "Create a worktree for feature/new-api, only check out apps/api and packages/core"

You:

1. Detect project: my-monorepo (from git remote)
2. Detect package manager: bun (found bun.lockb)
3. Allocate 2 ports: ~/.claude/skills/worktree-manager/scripts/allocate-ports.sh 2 → 8100 8101
4. Create worktree with sparse-checkout:

   mkdir -p ~/tmp/worktrees/my-monorepo
   git worktree add ~/tmp/worktrees/my-monorepo/feature-new-api -b feature/new-api --no-checkout
   ~/.claude/skills/worktree-manager/scripts/sparse-checkout.sh setup \
     ~/tmp/worktrees/my-monorepo/feature-new-api apps/api packages/core
   

5. Install deps (only in checked-out directories):

   (cd ~/tmp/worktrees/my-monorepo/feature-new-api && bun install)
   

6. Register with sparse-checkout info:

   ~/.claude/skills/worktree-manager/scripts/register.sh \
     "my-monorepo" "feature/new-api" "feature-new-api" \
     "$HOME/tmp/worktrees/my-monorepo/feature-new-api" \
     "/path/to/my-monorepo" "8100,8101" "Build new API endpoints" \
     true "apps/api,packages/core"
   

7. Launch agent
8. Report:

   Created worktree with sparse-checkout:
   
   | Branch | Ports | Sparse Dirs | Task |
   |--------|-------|-------------|------|
   | feature/new-api | 8100, 8101 | apps/api, packages/core | Build new API endpoints |
   
   Only 2 directories checked out (saving disk space and IDE indexing time).