🚀 What Is This?
A drop-in configuration layer for Claude Code that transforms it from a capable AI assistant into an enterprise-grade development powerhouse.
One git clone gives you 180 domain skills, 10 specialist agents, 81 community marketplaces with 11,000+ additional skills, 29 slash commands, 7 lifecycle hooks — all auto-activating based on what you're working on. No manual configuration required.
How it works: Describe what you want in plain language. Everything is installed and available, but only a small core of skills keeps its full description in Claude's context. The rest is listed by name and found through a generated index (INDEX.md, index/graph.json) and the toolkit-router skill, so a large toolkit costs almost nothing until something is needed. Language rules load only for matching files.
Runs in YOLO mode on purpose: bypassPermissions plus a Full Access+++ contract in CLAUDE.md (your request authorizes in-scope work; explicit "do not"s are absolute), backed by a guard hook that blocks a short list of footguns. Personal and machine-specific rules live in a gitignored local/ layer.
Cross-machine Git safety: a session-start banner flags repos that are behind, unpushed, dirty, or diverged. Session-end and daily maintenance hooks checkpoint all pending changes in configured owned repos, including unfinished work, then push through the configured runner when repository state permits. They stop for conflicts or possible secrets and report other push blockers. Marketplace clones are pulled at most daily and are never committed.
Who it's for: Developers who use Claude Code and want deeper domain expertise, structured workflows, and quality guardrails without manual setup.
Route Map
For agents (and people) looking for one thing: open the smallest file that answers, then stop. Grep indexes; don't read them whole.
Up: github.com/travisjneuman · Not here: per-project instructions belong in each project's own CLAUDE.md or AGENTS.md; personal and machine-specific rules live in the gitignored local/ layer.
| Step | Open | Answers |
|---|
| 1 | CLAUDE.md (always loaded) | How to work: autonomy, git, safety, browser access, model selection, review-based validation |
| 2 | INDEX.md: rg -n -i "<keyword>" ~/.claude/INDEX.md, or invoke the toolkit-router skill, which does this | Which skill, agent, command, rule, checklist, or marketplace repo fits |
| 3 | skills/<name>/SKILL.md, agents/<name>.md, rules/<lang>.md, docs/reference/<kind>/<name>.md | The procedure itself |
| Need / keywords | Open | Notes |
|---|
| skill, domain, specialist, "is there something for X" | INDEX.md, skills/MASTER_INDEX.md | Generated by scripts/generate-index.mjs; never hand-edit. Relationships: index/graph.json |
agent, subagent, agent-* | agents/README.md | Core agents live in agents/; other specialists are agent-<name> skills |
slash command, /name | docs/COMMANDS.md | Sources in commands/ |
| language or stack rules, stack guide, checklist, workflow reference, tooling | docs/reference/README.md | Lists every rules/ file (path-scoped) and every on-demand reference doc |
| hook, guard, secret scan, status line, session start/end | hooks/README.md | Git hooks for this repo: scripts/hooks/ |
| marketplace, community skills | index/marketplace-catalog.json, docs/MARKETPLACE-GUIDE.md | Catalog exists once marketplaces are pulled |
| setup, new machine, configuration, troubleshooting | |
Map
graph LR
R[README] --> C[CLAUDE.md]
R --> I[INDEX.md]
R --> D[docs/README]
D --> REF[docs/reference/README]
REF --> RU[rules/]
R --> SK[skills/README]
I --> SK
R --> AG[agents/README]
R --> CM[docs/COMMANDS.md]
R --> HK[hooks/README]
R --> SC[scripts/README]
R --> TP[templates/README]
R --> LE[local.example/README]
R --> WS[website/README]
Contents
This page: What Is This? · Route Map · Quick Start · What's Included · How It Works · Key Features · Configuration · Multi-Machine Setup · Platform Support · What's New · Documentation · Philosophy
Front doors
- CLAUDE.md — always-loaded constitution: autonomy, git, safety, review-based validation
- INDEX.md — generated index of skills, agents, commands, rules, docs, marketplaces
- docs/README.md — every guide, routed by need
- docs/reference/README.md — rules, stack guides, checklists, workflow and tooling references
- docs/COMMANDS.md — all slash commands by category (
commands/ holds only commands)
- skills/README.md — skill routes and category browse aid
- agents/README.md — core agents and how specialist agent skills work
- hooks/README.md — lifecycle hooks, token cost, guard behavior
- scripts/README.md — setup, maintenance, count and index scripts
- templates/README.md — component, project, and operational templates
- local.example/README.md — template for the private
local/ layer
Archive (historical, not routed by the index)
⚡ Quick Start
Use as GitHub Template (Recommended)
Click the green "Use this template" button at the top of this repo to create your own copy with a clean commit history. Then clone your new repo to ~/.claude/.
One-Line Install
Or clone directly if you prefer:
git clone https://github.com/travisjneuman/.claude.git ~/.claude
Initialize Marketplace Clones
git clone https://github.com/travisjneuman/.claude.git ~/.claude
cd ~/.claude
bash ~/.claude/scripts/init-marketplaces.sh
Start Claude Code
# Start Claude Code — the toolkit loads automatically
claude
That's it. Claude Code loads the toolkit from ~/.claude/ in every session.
📦 What's Included
| Component | Count | Description |
|---|
| Skills | 180 | Domain expertise modules (React, security, DevOps, finance, etc.) |
| Agents | 10 | Specialist subagents for focused tasks (code review, debugging, etc.) |
| Commands | 29 | Slash commands for common workflows |
| Marketplace Repos | 81 | Community skill repositories (11,000+ additional skills) |
| Hooks | 7 | Wired public hooks (session lifecycle, status line, safety guards; not git hooks or the dispatcher) |
| Templates | 17 | Project scaffolding and task templates |
| Rules | 10 | Stack-specific coding guardrails (TypeScript, Python, Go, Rust, etc.) plus web legal-exposure defaults |
| Category | Skills |
|---|
| Web Development | react-native, vue-development, svelte-development, flutter-development, pwa-development, electron-desktop, tauri-desktop, frontend-enhancer |
| Backend & API | api-design, graphql-expert, database-expert, microservices-architecture, event-driven-architecture, websockets-realtime, serverless-development |
| DevOps & Cloud | devops-cloud, monitoring-observability, performance-engineering, mobile-cicd |
| AI & Data | ai-ml-development, llm-app-development, data-science |
| Security | application-security, authentication-patterns |
| Mobile | android-development, ios-development, macos-native, kotlin-multiplatform, react-native, flutter-development |
| Quality | generic-code-reviewer, generic-feature-developer, generic-design-system, generic-ux-designer, test-specialist, tdd-workflow, debug-systematic, tech-debt-analyzer |
| Business | business-strategy, finance, marketing, sales, product-management, hr-talent, legal-compliance, risk-management, operations, innovation |
| Creative | brand-identity, graphic-design, video-production, audio-production, ui-animation, |
See MASTER_INDEX.md for the full listing with descriptions.
⚙️ How It Works
Dynamic Routing
Every session loads CLAUDE.md, the unscoped rules, and the skill listing. Only the core skills and agents carry full descriptions there; every other skill, command, checklist, and stack guide is listed by name only, so it costs almost nothing until it is used. For specialized work, the toolkit-router skill greps the generated INDEX.md for the domain and loads what fits, including the indexed marketplace skills:
Your prompt
│
▼
┌──────────────────────────────┐
│ Core skills (described) │ ← matched from their descriptions
│ + toolkit-router → INDEX.md │ ← grep by domain for everything else
└──────────┬───────────────────┘
│
┌────────┼────────┬────────────┐
▼ ▼ ▼ ▼
┌──────┐ ┌──────┐ ┌──────┐ ┌────────────┐
│Skills│ │Rules │ │Agents│ │ Marketplace│
│ │ │& Chk │ │ │ │ skills │
└──────┘ └──────┘ └──────┘ └────────────┘
Path-scoped rules load only when matching files are read.
Hook Lifecycle
Hooks run automatically; most cost zero tokens:
Session Start
├── Daily background pull (~/.claude, marketplaces, optional project repos)
├── Repo health banner (only when something needs attention)
└── Private local hook (optional, from local/)
Before Bash / Write / Edit
└── Guard: blocks root rm -rf, force push, reset --hard, secret-file writes, AI trailers
After File Edits
└── Secret scan (tells Claude if a credential-looking string was written)
After Each Response
└── Once a day per machine: usage tally + commit-everything sweep (background)
Session End
└── Commit all pending work in configured owned repos, then push when repository state permits; log conflicts, possible secrets, and push blockers
Status bar
└── model · effort · folder (branch) · context % · 5h/7d usage · cost
See hooks/README.md for the full hook reference.
🔑 Key Features
The toolkit aggregates 81 community skill repositories as ignored local clones in plugins/marketplaces/. All are read-only (fetch but never push). Skills span security (Trail of Bits), full-stack development, scientific computing, SAP/enterprise, Elixir, Terraform, creative writing, and more. Some marketplace repos are installed as plugins, making their agents, commands, and skills fully active in the routing system alongside built-in resources. Non-installed repos contribute discoverable skills via keyword search.
# Search marketplace skills
find ~/.claude/plugins/marketplaces -name "SKILL.md" | xargs grep -li "kubernetes"
# Update all marketplace repos
bash ~/.claude/_pull-all-repos.sh
See docs/MARKETPLACE-GUIDE.md for the full catalog.
The guard.js hook blocks dangerous commands before execution:
| Blocked Pattern | Risk |
|---|
rm -rf / / rm -rf ~ | Filesystem destruction |
git push --force | Overwrite remote history |
DROP TABLE / TRUNCATE TABLE | Database destruction |
git reset --hard origin | Discard all local changes |
git clean -fd | Delete untracked files |
curl ... | sh | Pipe-to-shell (RCE) |
npm publish | Accidental package publication |
chmod -R 777 | Overly permissive permissions |
docker system prune -a | Destroy all Docker resources |
Additional safety: guard.js also blocks writes to .env, credential, key, .ssh, .gnupg, .git/, and node_modules files, and commits carrying AI attribution trailers. secret-scan.js checks every file Claude writes for credential-looking strings and tells Claude if it finds one.
All hooks use a Node.js-based runner (hooks/run-hook.js) that resolves paths via os.homedir(). This avoids WSL path issues on Windows where ~ resolves to /root/ instead of the Windows user home.
On Windows, the runner explicitly prefers Git Bash (C:\Program Files\Git\usr\bin\bash.exe) over WSL bash to prevent drive mount failures on network drives.
| Platform | Shell | Notes |
|---|
| macOS | /bin/bash | Standard bash via Git or Homebrew |
| Linux | /bin/bash | Standard bash |
| Windows | Git Bash (preferred) | Avoids WSL drive mount errors |
For complex tasks, spawn multiple specialist agents to work in parallel:
- Code review from multiple angles — security, performance, tests
- Cross-layer features — frontend + backend + database agents
- Competing hypotheses — debug with parallel investigation
Agent types include architecture analysts, debugging specialists, performance optimizers, security auditors, and many more specialists that run as agent-* skills. See agents/README.md.
No MCP servers are configured by default, which keeps every session's context lean. Add one with claude mcp add when a project needs it; docs/MCP-SERVERS.md covers common options (Playwright, memory, databases, GitHub, Context7) and their token cost.
The toolkit maintains context between sessions automatically:
- Resume —
claude --continue / --resume reopen a previous conversation; auto memory carries preferences and project facts across sessions; /handoff writes an explicit handoff note when you want one
- Repo health —
session-start-repo-health.sh flags repos that need a pull, push, or reconciliation before you edit them
- Repo sync —
session-start-pull.sh fast-forward pulls in the background, at most once a day
🛠️ Configuration
Key Files
| File | Purpose |
|---|
CLAUDE.md | Core rules: autonomy, workflow, git, safety, code standards |
settings.json | Claude Code settings, hook registrations, permissions |
counts.json | Public counts, identity lists, source revisions/digests and snapshot date |
plugin.json | Plugin metadata |
Explicit count publication
Public toolkit resources and committed manifest marketplace skill bodies are distinct populations. Local and hosted website builds consume the same generated snapshots. Pulling sources or committing does not regenerate them; authorize one explicit source refresh, inspect its output/diff, and then publish. Pins and recorded dates label selected source snapshots, not latest-upstream guarantees. See Count pipeline for commands, definitions, profile/portfolio destinations and the separate DESK-only media contract.
Common Customizations
Add your own project directories to auto-pull:
cp ~/.claude/.env.example ~/.claude/.env.local
# Edit CUSTOM_PROJECT_DIRS="/path/to/your/projects"
Switch action mode between proactive (default) and conservative — see docs/reference/workflows/action-policy.md.
Disable a hook — edit settings.json and remove the hook entry.
🖥️ Multi-Machine Setup
The toolkit uses a two-layer architecture:
- This repo (
~/.claude/) — Shared configuration, cloned on every machine
- Machine-specific (
.env.local, known_marketplaces.json) — Gitignored, per-device
New Device Setup
git clone https://github.com/travisjneuman/.claude.git ~/.claude
cp ~/.claude/.env.example ~/.claude/.env.local
# Edit .env.local with machine-specific paths
bash ~/.claude/scripts/fix-marketplace-paths.sh
See docs/NEW-DEVICE-SETUP.md for the full walkthrough.
💻 Platform Support
| Platform | Status | Shell | Notes |
|---|
| macOS | Full support | bash/zsh | Notifications via osascript |
| Linux | Full support | bash | Notifications via notify-send |
| Windows 11 | Full support | Git Bash | Hooks use Node.js runner to avoid WSL issues |
| WSL | Works | bash | Native Linux behavior inside WSL |
Requirements: Node.js (comes with Claude Code), Git, bash (Git Bash on Windows).
🆕 What's New
September 2026 — Modernization for current Claude Code
- Lean always-on context: 23 core skills keep full descriptions; every other skill and command is
name-only (still installed and invocable) and discoverable through the generated INDEX.md / index/graph.json and the new toolkit-router skill. Language rules are path-scoped.
- Full Access+++ in
CLAUDE.md: rewritten, shorter constitution (autonomous execution, explicit prohibitions, zero testing (validate by review), pull-first git, no AI attribution). attribution settings disable commit/PR trailers at the source.
- Hooks fixed and trimmed: the Bash and write guards now actually work (they read an env var Claude Code never set); per-prompt git injection, per-turn session summaries, and GSD update checks are gone; the pull runs at most daily and never auto-commits or pushes; new status line.
- AGENTS.md support:
claude-md-and-agents-md so projects shared with Codex and other agents load their AGENTS.md.
- Private
local/ layer: gitignored rules, hooks, and host settings, plus a pre-commit public-safety gate. Template in local.example/.
- Marketplaces: manifest-only in
.gitmodules; clones are local, no_push, blocked from commits by pre-commit and pre-push checks. scripts/add-marketplace.sh <url> adds one.
- GSD removed: the
gsd-core marketplace is no longer registered, routed to, or counted; the old GSD tutorial is in archive/.
March 2026 — Comprehensive audit, new repos, full domain coverage
- 6 new marketplace repos:
blader/humanizer (11.4K stars), phuryn/pm-skills (8.3K stars, 100+ PM skills), Lum1104/Understand-Anything (6.6K stars), SawyerHood/dev-browser (4.9K stars), slavingia/skills (4.5K stars), millionco/expect (2.3K stars). Current manifest: 109 repos total.
- Marketplace clone cleanup: Fixed orphaned directories, standardized the
.gitmodules manifest, verified no_push protection on marketplace clones
- Full domain coverage: New skills and agents for blockchain/Web3, data engineering, embedded/IoT, edge computing, compliance, and more
- New stack rules: Added Go, Rust, Java/Kotlin, C/C++, Swift/iOS, Flutter/Dart rules — path-scoped for zero token cost when not relevant
- Count synchronization: All counts now accurate across README, plugin.json, counts.json (127 skills, 86 agents, 94 commands, 109 repos)
Earlier in March 2026:
- Marketplace routing: All repos now surface naturally by domain — routing coverage went from 23% to 100%
- CLAUDE.md condensed: Auto-routing table reduced from 72 to 29 rows, saving ~600 tokens per conversation
- start-task trimmed: 156 → 82 lines, removed duplication with CLAUDE.md
See CHANGELOG.md for the full history.
📚 Documentation
💡 Philosophy
This toolkit follows three core principles:
- Everything activates dynamically. Describe what you want — the system loads what's needed. No memorizing commands.
- Token efficiency over completeness. Domain content loads on-demand, not upfront. Unused skills cost zero tokens.
- Safety by default. Dangerous commands are blocked, secrets are scanned, protected files are guarded — all before you make a mistake.
Maintained by Travis Neuman · claude.travisjneuman.com · MIT License
