Skip to content

borghei/Claude-Skills / engineering/git-worktree-manager

git-worktree-manager

Unverified

Run parallel Git worktrees without collisions: a port block per worktree, .env sync, health checks for stale or dirty trees, and safe cleanup of merged ones. Use when several branches or agents need running dev servers at once.

Git Worktree Manager

Plain git worktree add/remove is the easy part. The agent uses this skill for what git does not do: give every worktree its own port block so dev servers and Docker services do not collide, carry .env files across, spot worktrees that are stale, dirty or tracking deleted branches, and remove merged ones without losing uncommitted work. It suits multi-agent setups where each agent or terminal owns one worktree.

Core Capabilities

  • Worktree lifecycle — create worktrees from new or existing branches with deterministic naming, copy .env files, install dependencies by lockfile detection, list with clean/dirty + ahead/behind status, and safely remove with uncommitted-change detection.
  • Port allocation — deterministic per-worktree assignment (base + index * stride), collision detection against running processes, persistent map in .worktree-ports.json, and Docker Compose override generation.
  • Multi-agent isolation — one branch per worktree, one agent per worktree, no shared state, conflict-free parallel execution, task-ID mapping for traceability.
  • Cleanup automation — stale detection by age, merged-branch detection for safe removal, dirty-state warnings, and bulk cleanup with safety confirmations.

When to Use

  • You need 2+ concurrent branches open with running dev servers.
  • You want isolated environments for feature work, hotfixes, and PR review.
  • Multiple AI agents need separate workspaces that do not interfere.
  • Your current branch is blocked but a hotfix is urgent.
  • You want automated cleanup instead of manual rm -rf operations.

Clarify First

Before managing worktrees, settle the inputs below. When one is missing or vague and would change the result, ask about it; when it wouldn't, use a sensible default and say which one you used.

  • Operation: create, list, remove, or clean up worktrees (selects the worktree_manager.py action vs port_allocator.py vs worktree_validator.py).
  • Branch & base: which branch(es) the worktree(s) track and from what base (drives worktree naming and isolation).
  • Port/service needs: whether dev servers or Docker need allocated ports (drives the deterministic port block and the Docker Compose override generated).

Ask about the two or three inputs that matter most for this request, not the whole list. If the user wants a result straight away, produce it and state your assumptions at the top.

Workflow

  1. Create the worktree with worktree_manager.py create <branch> --base main; it names the directory deterministically, copies .env files and installs dependencies from the lockfile it finds.
  2. Assign ports with port_allocator.py assign <worktree-name>: a block at base + index * stride, checked against listening processes and recorded in .worktree-ports.json. Use the Docker Compose override in references/setup-and-ports.md to apply it.
  3. Work: one branch and one agent per worktree, no shared mutable state.
  4. Check health with worktree_validator.py (stale trees, missing directories, deleted branches, .env parity, port conflicts, lockfile drift).
  5. Clean up with worktree_manager.py cleanup --dry-run first, then without --dry-run; dirty worktrees are reported, not deleted. Run port_allocator.py sync afterwards to free released blocks.
  6. Verify that worktree_validator.py reports no issues and git worktree list matches the port registry.

Quick Start

Run inside any git repository:

python3 scripts/worktree_manager.py list                          # worktrees with clean/dirty and ahead/behind
python3 scripts/port_allocator.py status                          # current port blocks
python3 scripts/worktree_validator.py --stale-days 14 --json      # health report
python3 scripts/worktree_manager.py cleanup --stale-days 30 --dry-run

Tools Overview

ToolSubcommands and flagsExit codes
worktree_manager.pylist, create <branch> [--base], remove <name>, cleanup [--stale-days N] [--dry-run]; --json0, 1 git operation failed, 2 no subcommand or not inside a repository
port_allocator.pystatus, assign <name> [--index] [--branch], release <name>, check, sync; --json0, 1 operation failed, 2 no subcommand or not inside a repository
worktree_validator.py--checks stale,missing,branch,env,ports,lockfile, --stale-days, --json0 no errors (warnings allowed), 1 errors found, 2 unknown check or not inside a repository

Anti-Patterns

Mistake: Deleting worktree directories with rm -rf. Why it happens: It looks like any other folder. Instead: Use worktree_manager.py remove (or git worktree remove); a raw delete leaves stale metadata and loses uncommitted work without warning.

Mistake: Two worktrees on the same dev-server ports. Why it happens: Every checkout reads the same default .env. Instead: Assign a port block per worktree and generate the Compose override from it.

Mistake: Two agents sharing one worktree. Why it happens: It saves a checkout. Instead: One agent per worktree and branch; shared working trees produce interleaved, unreviewable changes.

Files

  • scripts/worktree_manager.py, scripts/port_allocator.py, scripts/worktree_validator.py — the tools above. They read the current repository; there is no sample data.

Reference Documentation

Load the reference that matches the task — keep this file lean and pull detail on demand:

  • references/setup-and-ports.md — quick-start git worktree commands, the deterministic port-allocation strategy and .worktree-ports.json format, collision-detection snippet, the full setup-worktree.sh script, and the Docker Compose per-worktree override. Read when creating a worktree or wiring up ports/services.
  • references/cleanup-and-workflows.md — the cleanup-worktrees.sh safe-cleanup script, the multi-agent assignment pattern and rules, the scenario→action decision matrix, and the post-creation validation checklist. Read when automating cleanup or coordinating agents.
  • references/operations-playbook.md — common pitfalls, best practices, the troubleshooting table, and the success-criteria bar. Read before shipping a workflow or when diagnosing problems.

Scope & Limitations

This skill covers:

  • Git worktree lifecycle: creation, listing, status inspection, and removal
  • Deterministic port allocation and collision avoidance for parallel dev servers
  • Environment file synchronization and Docker Compose override patterns
  • Multi-agent workspace isolation strategies and cleanup automation

This skill does NOT cover:

  • Git branching strategies or merge conflict resolution (see pr-review-expert and release-manager)
  • Secret rotation, vault integration, or credential management (see env-secrets-manager)
  • CI/CD pipeline configuration or automated test orchestration (see ci-cd-pipeline-builder)
  • Monorepo package management, workspace linking, or cross-package dependency resolution (see monorepo-navigator)

Related skills: engineering/env-secrets-manager (the .env files copied into worktrees), engineering/monorepo-navigator (package boundaries inside a worktree).

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
discernment-nudgeanthropics/skillsAfter you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.Stars: 195kUpdated Claude CodeCodexVerifiedInstall
internal-commsanthropics/skillsA set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).Stars: 195kUpdated Claude CodeCodexVerifiedInstall