CC Commander — AI PM for Claude Code · 71 Skills · 22 Agents · Free for Now

Ship faster with Claude Code and Cowork Desktop. One plugin install activates 71 skills, 22 specialist AI agents, and 23 lifecycle hooks across Claude Code Desktop, Claude Code CLI, Cursor, Windsurf, and every MCP-capable IDE.

Install in 30 seconds:

Settings → Plugin Marketplace → Add from GitHub: KevinZai/commander → Install

What you get: 71 click-first /ccc-* plugin skills · 22 specialist sub-agents (architect, reviewer, debugger, designer, and more) · 23 lifecycle hooks · 2 bundled MCP servers · 502+ ecosystem skills · Free for now.

🖥️ Primary surface: Claude Code Desktop (aka Cowork Desktop). Install once via Settings → Plugin Marketplace. All 71 plugin skills, 22 agents, 23 lifecycle hooks (39 handlers), and 2 credential-free bundled MCP servers (+16 opt-in via /ccc-connect) appear inside every session automatically — no terminal needed.

Cowork Desktop and Claude Code Desktop are the same app, two UI modes. The plugin works identically in both. All screenshots in docs/screenshots/ were taken in Desktop.

Powered by: · affiliate disclosure

Kevin Zicherman · @kzic · Built from 200+ community sources · Aggregates 19 vendor packages

Why CC Commander · The 70 skills · Browse Skills · Agent Bible · Ecosystem · BIBLE · Changelog

🎯 Who it's for

If you are...	CC Commander is...
👋 New to AI coding agents → using Claude Cowork Desktop	Your onboarding buddy — /ccc pops up a native chip picker: Build · Review · Ship · Design · Learn · More. Click. No typing. 28 guided workflows, zero config.
💻 A developer → using Claude Code Desktop or Claude Code CLI	Your project manager — routes complexity, compounds knowledge across sessions, Kevin Z Method built in. /ccc-plan → spec interview → plan file.
🔧 In Cursor / Windsurf / Cline / Continue / Codex	One hosted MCP endpoint unlocks all 459+ skills in your IDE of choice
📱 On Claude mobile or iPad	Same skills, same license, synced across devices (hosted MCP)

---

⭐ 30-second install

For Cowork Desktop or Claude Code Desktop users (99% of people). No terminal needed.

1. Open Settings → Plugin Marketplace
2. Click Add from GitHub → enter KevinZai/commander → Add
3. Find commander in the marketplace → click Install
4. Cmd+Q, reopen the app, type /ccc — you're live

✅ 71 plugin skills activate. ✅ Zero config. ✅ Zero API keys.

Power-user CLI install:

/plugin marketplace add KevinZai/commander
/plugin install commander

Or full dev environment + ccc binary:

curl -fsSL https://raw.githubusercontent.com/KevinZai/commander/main/install-remote.sh | bash

---

🌐 One plugin, every agent

Commander is a PM layer for whichever AI coding tool you use. Same 71 skills, same 22 specialist agents, same lifecycle hooks — different install path per platform. The package name stays commander everywhere; the surrounding UI (Claude Code, Codex, Cursor, etc.) provides the platform context.

Platform	Status	Install
Claude Code Desktop / Cowork Desktop ⭐	Shipping (primary)	Settings → Plugin Marketplace → Add from GitHub: KevinZai/commander
Claude Code CLI	Shipping	/plugin marketplace add KevinZai/commander then /plugin install commander
Cursor / Windsurf / Cline / Continue / Codex	Shipping (hosted MCP)	Point MCP client at mcp.commanderplugin.com
Codex / Gemini native plugin runtimes	Roadmap	Same repo, platform-specific adapter (planned v4.2+)

One brand: Commander. One tagline: The guided AI PM for every AI coding agent. One repo: KevinZai/commander. Everywhere.

---

🎯 Why CC Commander

Stock Claude Code is a blank terminal with amnesia. Every other plugin solves one slice. CC Commander is the first guided AI PM — click-first UX, an intelligence layer that thinks before you do, and the whole ecosystem pre-wired. One install, everything active.

🥇 What no competitor has

Moat	CC Commander	Cursor / Windsurf / Copilot	OSS (Cline / Continue / Aider)
Lifecycle hooks (23 × 39 handlers)	✅ SessionStart, PreToolUse, PostToolUse, Stop, PreCompact, Notification, etc.	❌ None	❌ None
22 specialist agent personas	✅ Architect, Security-Auditor, TypeScript-Reviewer, Designer, Debugger, etc.	❌ None ship pre-built	❌ None ship pre-built
18 MCP servers (2 bundled + 16 opt-in)	✅ Zero-config — Notion, Slack, Supabase, Figma, Linear, GitHub, Vercel, Fly…	⚠️ Manual MCP setup required	⚠️ Manual MCP setup required
Spec-first planning (/ccc-plan → interview → spec → multi-agent execution)	✅ Native	❌ No planning primitive	❌ No planning primitive
Cross-platform (CLI + Desktop + VS Code path + hosted MCP)	✅ Same brain everywhere	❌ IDE-locked	⚠️ IDE-locked
Pricing transparency	✅ Free for now — no signup, no card, no gates	⚠️ Cursor $20+ confusing tiers; Windsurf credit confusion	✅ BYOK free

The marketing angle: Set it once, the AI works smarter every session. Lifecycle hooks fire automatically. Specialist agents already know your codebase conventions. The MCP ecosystem is pre-wired. No other tool ships any of this out of the box.

• 🖱️ Click-first UX — every menu is a native AskUserQuestion chip picker. No typing. No numbered menus. No ASCII prompts. Works identically in Cowork Desktop, Claude Code Desktop, and the CLI.
• 🧠 /ccc-suggest intelligence layer — Opus-class real-time recommendation. Scans your project state, recommends one starred next step with reasoning + named 3rd-party plugins. Kills info-paralysis.
• 🧩 71 plugin skills including 13 /ccc-* specialist workflows — plain slash commands (no commander: prefix), skill-based architecture: ccc, ccc-start, ccc-browse, ccc-plan, ccc-build, ccc-review, ccc-ship, ccc-design, ccc-learn, ccc-xray, ccc-linear, ccc-fleet, ccc-connect + /ccc-e2e, /ccc-save-session, /ccc-resume-session, /ccc-changelog, /ccc-doctor, /ccc-tuneup, /ccc-upgrade, /ccc-yolo-setup, /ccc-ecc and 11 domain routers.
• 🧭 Orchestrator / Executor model — /ccc-orchestrate has Fable/Opus write a Skill.md-style goal file, GPT-5.5 via codex or a Sonnet subagent execute it, and the orchestrator verify acceptance criteria. Pair it with /ccc-handoff for frequent fresh-chat resets and /ccc-adopt to bring the doctrine into any existing project. Pay for Fable on the thinking, not the typing.
• 🎭 22 specialist agents with persona voices — architect, security-auditor, performance-engineer, content-strategist, data-analyst, designer, product-manager, technical-writer, devops-engineer, qa-engineer, reviewer, builder, researcher, debugger, fleet-worker, typescript-reviewer, python-reviewer, go-reviewer, rust-reviewer, java-reviewer, kotlin-reviewer, csharp-reviewer. Each with a distinct voice layer in commander/cowork-plugin/rules/personas/.
• 🔌 2 credential-free bundled MCP servers — context7 (library docs) + sequential-thinking. Opt-in via /ccc-connect: 16 more including Tavily, GitHub, Supabase, Figma, Playwright, Slack, Notion, claude-mem, Exa, Firecrawl, Zapier, Google Drive, Vercel, Neon, Fly.io, Upstash.
• 🪝 23 lifecycle hooks × 39 handlers — SessionStart, UserPromptSubmit, PreToolUse, PostToolUse, Stop, Notification, PreCompact, SubagentStop, PermissionRequest. Schema-compliant, 100% pass on claude plugin validate.
• 🗺️ /ccc-cheatsheet live Mermaid map — reads the filesystem as single source of truth, renders the whole plugin as a flow diagram. Never drifts.
• 🔄 Weekly vendor auto-sync — 19 vendor submodules auto-update via GitHub Actions. The ecosystem stays current without you touching a thing.
• 🌐 Cross-client, one license — Cowork Desktop, Code Desktop, Code CLI, Cursor, Windsurf, Cline, Continue, Codex, mobile (hosted MCP).
• 🚁 Fleet orchestration — /ccc-fleet runs multiple Sonnet agents in parallel git worktrees. Fan-out, pipeline, FOR/AGAINST, background modes.
• 🆓 Free for now — no credit card, no signup. No feature gating, no paywalls.
• 🧬 Shared brain: plugin + CLI — same intelligence layer, same skill catalog, same personas. Install either, get both.
• 📖 The Kevin Z Method — BIBLE.md ships with the plugin. 7 rules, 200+ sources, 14 months of production methodology.
• 📄 Printable cheat sheets — docs/cheatsheets/ ships 3 print-ready PDFs: CC Commander Essentials, The Orchestrator/Executor Method, and Agents & Model Tiers.
• 📚 459+ skills across 11 CCC domains — design, marketing, SaaS, DevOps, SEO, testing, security, data, research, mobile, makeover. Plus the vendor ecosystem.
• 🎯 Plugins-name-plugins — /ccc-suggest calls out specific 3rd-party plugins by name (claude-mem, superpowers, caveman, impeccable, graphify, and more) at the right moment. You learn the whole ecosystem through CC Commander.

🧠 The Fable Method

The operating doctrine distilled from Anthropic's frontier Fable 5 model, encoded as 12 enforceable gates — so ANY model running CC Commander produces Fable-shaped results, not just the flagship one. Arm it with /ccc-fable on, self-check with /ccc-fable audit. Full doctrine: commander/cowork-plugin/rules/fable-method.md.

• Never trust a single pass — the agent that does the work never grades it. Every finding gets adversarial verification from a fresh context.
• Loops need verifiers, state, and stops — repetition without a real verifier, a state file, and a deterministic stop condition is just an agent agreeing with itself on repeat.
• Prove it before you alarm — a scary number is a hypothesis until you run the exact check, open real samples, and hunt the disconfirming case.
• Context is disposable, state is durable — the durable record is handoff docs, memory, and the tracker, never the live chat.

The method IS the moat. A mid-tier model following all 12 gates reliably out-delivers an ungated frontier model on any task longer than one turn.

💡 How the intelligence works (the beginner headline)

New to Claude Code? Start here: /ccc-suggest.

Every time you run it, CC Commander reads your project state — detected stack, current branch, last 5 commits, open todos, recent errors — and returns one starred recommendation with plain-English reasoning. If another plugin would serve you better right now (e.g. claude-mem for persistent memory, superpowers for structured thinking, caveman for token-heavy iteration), it names it. Three reasoning tiers fire in order:

1. Strong signals — an open PR, a failing test, a dirty worktree, a mid-plan session
2. Stack signals — nextjs + Stripe detected → suggest ccc-saas; mobile repo → suggest ccc-mobile
3. User intent — pattern-match your recent prompts against the 502-skill catalog

One click → one next step. No info paralysis.

---

Built on Claude Agent SDK primitives

CC Commander is built on Anthropic's 2026 Claude Agent SDK sub-agent architecture — the same primitives that power agentic workflows across Claude Code Desktop, Cowork Desktop, and the CLI. The brain/hands pattern separates orchestration (the PM layer: planning, routing, decision-making) from execution (22 specialist sub-agent personas that each embody a distinct role, model, and voice). 23 lifecycle hook events with 39 handlers fire automatically throughout every session — from SessionStart initialization through SubagentStop result aggregation — so the right agent is always running at the right moment without you lifting a finger.

---

Using Cursor, Windsurf, Cline, Continue, or Codex?

Hosted MCP setup — one URL + license key unlocks all 459+ skills in your editor.

---

🌟 What's new in v5.1.0

• 🧠 Opus 4.8 session default — claude-opus-4-8, $5/$25 per MTok, Fast mode 2.5× at $10/$50. New ultra/xhigh effort levels. thinking: {type: "adaptive"} replaces deprecated budget_tokens.
• ⚡ Dynamic workflows (research preview, requires Claude Code v2.1.154+) — 4 bundled JS workflow scripts spawn parallel subagents that cross-check findings: ccc-audit (backs /ccc-xray), ccc-deep-review (backs /ccc-review), ccc-migrate, ccc-fleet. Skills fall back to Agent() on older clients.
• 🚀 /ccc-ultracode — new skill for xhigh effort + automatic workflow orchestration. Or add workflow: to any prompt for a one-off run.
• 🪝 Hooks 9 → 23 events — 14 new lifecycle events: SessionEnd, PostCompact, SubagentStart, TaskCreated, TaskCompleted, and 9 more. Handlers: 24 → 38. Correctness fix: session-save moved Stop → SessionEnd.
• 🎯 71 plugin skills — recent channel/CI/ECC/setup skills (/ccc-brainstorm, /ccc-qa, /ccc-loop, /ccc-hermes, /ccc-nightwatch, /ccc-ci, /ccc-yolo-setup, /ccc-ecc) + 5 language reviewer agents (Go, Rust, Java, Kotlin, C#) + /ccc-ultracode.
• 🔐 4 security mediums resolved — telemetry scrubber hardened, error messages sanitized.
• 🤖 22 specialist sub-agent personas — 5 new language reviewers added (Go · Rust · Java · Kotlin · C#).
• 📦 15 vendor pins refreshed — ECC v1.7.0, oh-my-claudecode v4.14.4, repomix v1.14.1, and 12 more.
• 🧪 284 tests passing — +5 security regression tests added.
• 🖥️ Desktop-first positioning throughout — all docs now lead with "Claude Code Desktop is the primary surface."

🚀 What's new in v6.0.0

• 🧠 Selective Fable 5 deep-mode — Opus 4.8 stays session default (everyday), Fable 5 escalation-tier via /model claude-fable-5 with once-per-day deep-reasoning nudge. Motto: pay for Fable on the thinking, not the typing.
• 🎛️ Smart model routing — selectModelForComplexity(score) auto-assigns: 0-29 Haiku, 30-65 Sonnet, 66-85 Opus, 86-100 Fable. Dispatch tiers: power=Fable/Opus, assisted=Opus/Sonnet, guided=Sonnet/Haiku.
• 💰 Savings counter + ccc --savings — every dispatch logs estimated cost vs baseline. Footer shows daily estimate (sv$X.XX); report command prints full breakdown to ~/.claude/commander/savings.json.
• 🎭 4 deep personas on Fable — architect, debugger, security-auditor, product-manager run claude-fable-5 for complex reasoning.

⚡ Dynamic Workflows + Ultracode

Research preview — requires Claude Code v2.1.154+. Skills fall back to standard Agent() dispatch on older clients.

Dynamic workflows are JS scripts the runtime executes in the background, spawning many subagents that cross-check findings and merge results before returning. Four bundled workflows ship in commander/cowork-plugin/workflows/:

Workflow	Backs	What it does
ccc-audit	/ccc-xray	Repo-wide audit — security, performance, architecture, test coverage in parallel
ccc-deep-review	/ccc-review	4-dimension branch review (correctness · security · perf · maintainability) + reconciler
ccc-migrate	/ccc-build migrate	Discover → transform → verify pipeline for safe large migrations
ccc-fleet	/ccc-fleet	Fan-out / pipeline / judge orchestration patterns

Ultracode combines xhigh effort + automatic workflow orchestration. Three ways to activate:

/ccc-ultracode               # guided path — recommended
/effort ultracode            # set for current session
workflow: <your task here>   # one-off without mode switch

The built-in /deep-research is also available for multi-source validated research.

The 71 plugin skills

Each /ccc-* workflow works via native Desktop chip picker — no typing, just click:

Command	What it does	First click
/ccc	Main hub — 6 intent tiles	Build · Review · Ship · Design · Learn · More
/ccc-suggest 🌟	Intelligence layer — recommends ONE starred next step	Scans state → 1 move + reasoning + named plugins
/ccc-start	First-run onboarding + plan file	New project · Existing · Tour · Skip
/ccc-browse	Searchable catalog of every skill + agent	Domains · Workflows · Agents · All
/ccc-plan	Spec interview → implementation plan file	New feature · Bug fix · Refactor · From Linear
/ccc-build	Scaffold a project	Web app · API · CLI · Mobile · From spec
/ccc-review	Audit current branch	Diff · Security · Performance · Full x-ray
/ccc-ship	Pre-flight + deploy	Pre-flight · Release · Deploy · Rollback
/ccc-design	UI/UX workflow	Landing · Components · Polish · Figma→code
/ccc-learn	Skill discovery	Design · Marketing · SaaS · More domains
/ccc-xray	Project health scorecard	Quick · Full · Deps-focused · Security-focused
/ccc-linear	Linear board integration	View · Pick · Create · Overview
/ccc-fleet	Multi-agent orchestration (git worktrees)	Fan-out · Pipeline · FOR/AGAINST · Background
/ccc-connect	Opt-in MCP connector	Notion · Zapier · Supabase · Slack · GDrive
/ccc-cheatsheet	Live Mermaid map of the whole plugin	Filesystem is single source of truth
/ccc-e2e	E2E test scaffolding + Playwright automation	New suite · Existing project · Visual regression
/ccc-save-session	Save session state for later resume	Saves context, todos, plan to ~/.claude/commander/
/ccc-resume-session	Resume a previously saved session	Lists recent sessions → pick one to restore
/ccc-orchestrate	Cross-runtime Orchestrator/Executor	Fable/Opus goal file → GPT-5.5 or Sonnet executor → verified acceptance criteria
/ccc-handoff	Proactive fresh-chat handoff	Dense session file + explicit restart before context quality decays
/ccc-adopt	Adopt CCC doctrine in another repo	Merge Orchestrator/Executor markers into CLAUDE.md without clobbering project rules
/ccc-fable	Arm the Fable Method — 12-gate session contract	on · status · audit · off — model-agnostic
/ccc-yolo-setup	Safe-YOLO permissions + Plan mode guardrails	Conservative · Balanced · Aggressive-safe
/ccc-ecc	Selective ECC component loader	Load skill · Load agent · Load hook · List

Plus 11 domain routers (ccc-design, ccc-marketing, ccc-saas, ccc-devops, ccc-seo, ccc-testing, ccc-security, ccc-data, ccc-research, ccc-mobile, ccc-makeover) + 2 vendor-sourced skills (/ccc-agent-writing, /ccc-systematic-debugging). See SKILLS-INDEX.md for the full catalog.

Click a cell → pick a sub-option → CC Commander handles it. No config files, no YAML editing, no syntax to memorize.

🤝 Works with these amazing OSS plugins

/ccc-suggest doesn't hoard — when another plugin is the right tool, it names it. Here's what CC Commander recommends at the right moment:

Plugin	When /ccc-suggest calls it out
claude-mem	Persistent cross-session memory — you're 10+ sessions into a project and want recall
superpowers	Forces structured thinking — you're drifting on a multi-step task
caveman	Token-heavy iteration — cuts ~75% output tokens during rapid loops
impeccable	Final polish pass on UI — pixel-perfect micro-interactions
graphify	Any input (code, docs, papers) → clustered knowledge graph + HTML report
claude-reflect	Self-improving skills with reflection loops — you keep repeating a mistake
Everything Claude Code (ECC)	Full harness — 156 skills, 72 commands, 38 agents, lifecycle hooks
gstack	CEO/eng plan review, office hours, QA gates
Compound Engineering	Knowledge compounding + mandatory code review enforcement
repomix	Pack a codebase for AI (60% smaller via tree-sitter compression)
claude-hud	Real-time status display + offline cost tracking + git diffs
RTK	Token optimization (60–90% savings) + 25 AWS subcommands
oh-my-claudecode	HUD with worktree support, quota tracking, hyperlinks
acpx	ACP protocol, Flows system, structured agent-to-agent comms
Caliber	Config scoring, drift detection

Install any of these separately — CC Commander detects what's available and routes to it. If you've got claude-mem installed, /ccc-suggest will reach for it when memory recall is the right move.

🧩 How Commander Extends ECC

Everything Claude Code (ECC) is the comprehensive harness — 156 skills, 72 commands, 38 agents, full lifecycle hooks. CC Commander is not a replacement for it — it's the PM/orchestration UX layer on top.

Layer	Owns
ECC (harness)	Raw breadth — skills, commands, agents, hook lifecycle
CC Commander (UX)	Curation + guidance + memory: click-first /ccc-* chips (AskUserQuestion), /ccc-suggest routing, the brain/hands pattern, The Kevin Z Method

What Commander adds on top of ECC:

• Click-first front door — 70 curated /ccc-* workflows via native chip pickers, instead of memorizing 156+ commands.
• /ccc-suggest — Opus-class "next best move" routing across the whole ecosystem (names ECC when ECC is the right tool).
• /ccc-orchestrate + /ccc-handoff + /ccc-adopt — cross-runtime plan/execute split, proactive context reset, and one-command doctrine adoption for existing projects.
• /ccc-ecc — selective ECC loader for one skill, agent, or hook without installing the full harness.
• Memory + compounding — sessions learn; /ccc-knowledge + claude-mem integration.
• Vendor aggregation — ECC is one of 19 vendor packages; the Smart Orchestrator scores and routes across all of them.

ECC-sourced parts are labeled in the vendor tree (vendor/everything-claude-code/). Commander ports a minimal hook/agent set and otherwise defers to ECC upstream — run /ccc-upgrade to bump the pin. Install ECC separately for the full harness; Commander detects it and routes to it.

🆚 Compare to other Claude Code plugins

CC Commander is a meta-layer — it routes to other plugins, not away from them. Here's how it differs from the top alternatives:

	CC Commander	ECC	Superpowers	gstack
Primary surface	Claude Code Desktop plugin + CLI	CLI harness	Claude Code CLI	Claude Code CLI
Click-first UX	✅ Native chip pickers, no typing	❌ Text menus	❌ Slash commands	❌ Slash commands
Sub-agent personas	22 (architect, reviewer, debugger…)	38 agents	0	3
Skill count	502+ (71 plugin + 450+ ecosystem)	156	~20	~15
Hosted MCP server	✅ (100 calls/mo free)	❌	❌	❌
Cross-IDE	✅ (Cursor, Windsurf, Cline, Codex)	❌ Claude only	❌ Claude only	❌ Claude only
Free	✅ Forever	✅	✅	✅
Recommends rivals	✅ /ccc-suggest names other plugins	❌	❌	❌

Bottom line: if you want the full Claude Code harness (156 skills, 72 commands), install ECC. If you want the click-first PM layer that works across Cowork Desktop, CLI, and Cursor, install CC Commander. Many users run both.

2-Minute First Win

# 1. Open the command center
/ccc

# 2. Say: "build me a REST API with TypeScript"
# → Commander asks 3 clarifying questions
# → Selects the right model + skills automatically
# → Ships working code with tests

# 3. Say: "review my last PR"
# → reviewer agent runs severity-rated code review
# → Flags Critical/High/Medium/Low findings with file:line citations

# 4. Say: "night mode — build a landing page while I sleep"
# → 8-hour autonomous build with checkpoints every 10 edits
# → Wakes you up to shipped code

Works In 8 IDEs

Claude Code · Cursor · Windsurf · Cline · Continue · Codex · Claude mobile · Claude Desktop

Free for Now

All 71 plugin skills, 22 specialist agents, 23 lifecycle hooks, 2 bundled MCP servers, and 16 opt-in connectors — free, no credit card, no feature gating.

---

---

Why CCC

Stock Claude Code is a blank terminal with amnesia. No skills. No guidance. No memory. Every session starts from zero. And it wastes 98% of your context window on tool output you'll never re-read.

CCC remembers everything, learns from every session, and gets smarter the more you use it. It wraps every major Claude Code tool into one install — with a smart orchestrator, guided menus, and an Intelligence Layer that auto-adjusts every dispatch based on your project.

You type: ccc
You get:  A guided AI project manager with 502+ skills,
          71 plugin skills, 22 agents, 19 vendor packages,
          real learning, and zero setup.

---

See It In Action

Main Menu — Arrow Keys, No Commands to Memorize

Cockpit Dashboard — Live Session Status

Context, rate limits, and budget meters in your terminal. Color-codes green → yellow → red as limits approach.

Rich footer bar — rainbow status line with live session meters:

━━ CCC5.1│🔥Opus4.8-1M│🔑gAA│🧠▐██45%░░▌│⏱️▐██░░▌6%│📅▐██░░▌34%│💰$2.34│⬆️640K⬇️694K│⏰8h0m│🎯502│📋CC-150│📂~/project

Stats Dashboard

Sessions, streaks, badges, cost tracking, activity heatmap, level progression.

11 CCC Domains — One Skill Loads an Entire Specialty

500+ Skills — Install Only What You Need

All recordings are real terminal output captured with vhs. No mockups.

---

📸 Screenshots

Desktop screenshots arriving with the v5.1.0 stable tag — see docs/screenshots/PLACEHOLDERS.md for in-progress capture briefs.

---

Intelligence Layer

Stock Claude Code is a blank terminal with amnesia. CCC remembers everything, learns from every session, and gets smarter the more you use it.

An AI project manager that thinks before it acts. Before dispatching a single token, CCC scores your task, reads your stack, pulls relevant lessons from past sessions, and selects the right model and budget automatically. No configuration. No flags. It just works.

How It Works

Four modules. Always running.

1. Weighted Complexity Scoring (dispatcher.js)

Every task scored 0–100 using 47 keyword signals, word count, and fuzzy regex matching:

"fix typo"             → score  0  → 10 turns, $1 budget, Haiku
"add dark mode"        → score 25  → 20 turns, $3 budget, Sonnet
"refactor auth module" → score 60  → 35 turns, $6 budget, Sonnet
"build SaaS platform"  → score 100 → 50 turns, $10 budget, Opus

File scope estimation adds 0–20 bonus points by scanning how many project files the task is likely to touch. Budget and turns auto-adjust — no manual flags needed.

2. Stack Detection (project-importer.js)

CCC reads your project before every dispatch: package.json, Dockerfile, go.mod, requirements.txt. Detects nextjs, react, vue, docker, python, rust, go, github-actions, orm, billing, testing. Reads current git branch + last 5 commit themes. Adds monorepo detection (workspaces, lerna, turbo, nx).

3. Skill Recommendations (skill-browser.js)

recommendSkills(task, techStack) combines three signals:

Signal	Weight	What It Does
Stack match	10 pts	Next.js project → nextjs-app-router ranks first
Keyword match	2 pts/hit	"auth" task → auth, jwt, better-auth bubble up
Usage history	boost	Skills that worked for you rank higher over time

Trending skills (7-day window) surface automatically. Skills that led to successful sessions compound their ranking advantage.

4. Knowledge Compounding (knowledge.js)

Every completed session extracts a lesson (keywords, category, stack, error patterns, success patterns) stored in ~/.claude/commander/. Searched before the next dispatch with time-decay relevance: < 7 days = 2x, < 30 days = 1.5x, older = 1x. Fuzzy keyword matching and cross-domain boosts (web↔react, api↔backend, testing↔bugfix) catch related concepts.

Smart retry handles failures automatically: rate limit → wait 60s + retry; context overflow → reduce turns to 60% + retry; budget exceeded → clear error with next steps.

Intelligence Analysis — Smart Dispatch in Action

Before dispatching a single token, CCC scores your task, reads your stack, and auto-configures the session.

  Task: "Add authentication with JWT and OAuth"

  Intelligence Analysis:
  ├─ Complexity Score: 72/100 (complex)
  ├─ Keyword signals:  +15 (auth) +15 (implement) +20 (production)
  ├─ Word count:       +10 (detailed description)
  ├─ Stack detected:   nextjs, react, tailwind
  └─ Related lessons:  2 found (JWT auth succeeded, OAuth2 pattern)

  Auto-configured dispatch:
     Model: opus  |  Turns: 35  |  Budget: $6
     Skills: nextjs-app-router, ccc-saas, auth-patterns
     Knowledge: 2 past lessons injected

  Dispatching... ████████████████████░░░░ 78%

Skill Recommendations — Ranked by Relevance

CCC ranks skills using your stack + task keywords + past usage. The right tools surface automatically.

  Recommended skills for your project:

  SCORE  SKILL                 WHY
  ─────────────────────────────────────────────
   20    nextjs-app-router     Stack match + keyword
   16    frontend-design       Stack match
   16    shadcn-ui             Stack match
   14    saas-scaffolder       Keyword: auth, billing
   12    tailwind-v4           Stack match
    8    ccc-testing           Always recommended

  ❯ Use top recommendation
    Browse all 459+ skills
    Search by keyword
    Back to main menu

Knowledge Compounding — Learning After Every Session

Every completed session extracts patterns and errors. The next dispatch is informed by everything that worked before.

  Session Complete — Knowledge Extracted

  Task:     "Add JWT authentication"
  Outcome:  Success
  Cost:     $4.12  |  Duration: 12m
  Category: api

  Patterns learned:
  ├─ bcrypt + JWT combo works well with Next.js
  ├─ httpOnly cookies for token storage
  ├─ middleware.ts matcher needs exact paths
  └─ next-auth conflicts with custom Prisma adapter

  Knowledge base: 47 lessons  |  12 categories
  Next dispatch will be 23% more informed.

Project Import — Stack Auto-Detection

CCC reads your project before every dispatch. No setup needed.

  Importing: ~/ccc-projects/my-saas-app

  Detected:
  ├─ CLAUDE.md           142 lines
  ├─ Tech stack:         nextjs, react, tailwind, prisma, stripe
  ├─ Monorepo:           No
  ├─ Git branch:         feature/checkout
  └─ Recent commits:     "add cart", "fix price calc", "stripe webhook"

  6 skills pre-selected for this stack
  3 relevant lessons from knowledge base

  ❯ Start building in this project
    Run /xray health audit first
    Back to main menu

The Net Effect

Session	What Changed
1	Dispatches based on complexity score alone
5	Knows your stack, recommends proven skills
20	Has learned your patterns — feels like a PM who knows your codebase

Token Optimization Stack — 5 Layers of Savings

Layer	Tool	Savings
Tool output sandboxing	context-mode	98% — SQLite + FTS5, BM25 snippets only
CLI output filtering	RTK	99.5% — strips verbose shell output
Skill tiering	_tiers.json	~10k tokens — 30 essential vs 459 full
Rate limit rotation	ClaudeSwap	2 MAX accounts, drain-first strategy
Prompt caching	Extended TTL	90% discount, 1hr cache window

---

Features

Component	Count	What It Does
Skills	502+	On-demand expertise (deduplicated)
Plugin Skills	71	Desktop-first skills (13 /ccc-* workflows + 11 domain routers + 7 new channel/ECC/setup skills + Orchestrator/Executor + meta + vendor-sourced + session management)
Agents	22	Specialist sub-agent personas with distinct voice layers
Lifecycle Hooks	23 × 39 handlers	SessionStart, UserPromptSubmit, PreToolUse, PostToolUse, Stop, Notification, PreCompact, SubagentStop, PermissionRequest
MCP Servers	2 + 16 opt-in	Context7 + Sequential-Thinking bundled; 16 more opt-in via /ccc-connect (GitHub, Supabase, Figma, Playwright, Slack, Notion…)
CCC Domains	11	Domain routers with sub-skills
Commands	83	Slash commands (/ccc- prefix)
Hooks (JS)	25	Lifecycle automation (kit-native)
Adventures	13	Guided interactive flows
Vendor Packages	19	Best-in-class tools, auto-updated
Themes	10	Cyberpunk, Fire, Ocean, Aurora, Sunset, Monochrome, Rainbow, Dracula + more
Prompts	36+	Battle-tested templates
Modes	9	Workflow presets
Split Mode	tmux	Tabbed sessions — each task gets a window
Agent API	CLI	Headless dispatch for AI orchestrators
Infrastructure	6	Fleet, Synapse, Cost, AO, CloudCLI, Paperclip commands
Service Detector	auto	Probes 8 services + 4 CLIs on startup

Every Path, Visualized

Every menu option, every sub-flow, every path to Claude Code — one diagram. View full size

Night Mode / YOLO — Autonomous Overnight Build

Start before bed. Wake up to shipped code.

10 questions → Opus with max effort → $10 budget → 100 turns → self-testing loop.

  YOLO Mode — Autonomous Overnight Build

  What are we building tonight?
  > "Full e-commerce checkout with Stripe integration"

  Configuration:
  ├─ Model:       Opus (max reasoning)
  ├─ Budget:      $10 (hard cap)
  ├─ Max turns:   100
  ├─ Self-test:   Enabled (tests after each phase)
  ├─ Checkpoint:  Every 10 edits
  └─ Stop file:   ~/.claude/commander/yolo-stop

  Ready to run for up to 8 hours unattended.

  ❯ Start YOLO build
    Adjust settings
    Cancel

---

11 CCC Domains

Each domain is a router that dispatches to specialized sub-skills on demand.

Domain	Skills	What's Inside
ccc-design	41	landing pages, UI audit, animation, responsive layout, color systems, typography, canvas design, wireframes, component library, accessibility, dark mode, micro-interactions, illustration, icon sets, design tokens, named-look design systems, video/HTML design capture
ccc-marketing	45	CRO, email campaigns, ad copy, social media, SEO content, blog posts, landing page copy, A/B testing, funnel optimization, lead magnets, newsletter, brand voice, press releases, case studies, video scripts
ccc-saas	20	auth systems, billing/Stripe, API design, database schema, multi-tenancy, onboarding flows, admin dashboards, role-based access, webhooks, rate limiting, usage tracking, feature flags
ccc-devops	20	GitHub Actions, Docker, AWS deploy, Terraform, monitoring, logging, CI/CD pipelines, Kubernetes, Nginx, SSL certs, environment management, health checks, rollback strategies
ccc-seo	19	meta tags, JSON-LD schema, sitemap, robots.txt, Core Web Vitals, internal linking, keyword research, content optimization, image SEO, page speed, structured data, canonical URLs
ccc-testing	15	Vitest, Playwright E2E, TDD workflow, snapshot testing, API testing, load testing, coverage reports, test fixtures, mock strategies, visual regression, accessibility testing
ccc-makeover	3	/xray project audit (health score 0-100, maturity 1-5), /makeover agent swarm execution, before/after report card
ccc-data	8	SQL optimization, data pipelines, analytics setup, data visualization, machine learning, reporting, data quality, vector search
ccc-security	8	OWASP top 10, secrets scanning, dependency audit, container security, penetration testing, CSP headers, rate limiting, auth hardening
ccc-research	8	competitive analysis, market research, user research, technology evaluation, trend analysis, SWOT, stakeholder interviews, data synthesis
ccc-mobile	8	React Native, Expo, mobile UI, push notifications, deep linking, app store optimization, offline-first, gesture handling

---

19 Vendor Packages

CCC aggregates the best Claude Code tools as git submodules. Auto-updated weekly. 19 packages, 1,500+ vendor skills.

Package	Stars	What You Get
Everything Claude Code	120K+	156 skills, 72 commands, 38 agents, lifecycle hooks
gstack	58K+	CEO/eng plan review, office hours, QA — OpenClaw integration v2
Superpowers	29K+	Forces structured thinking — /ccc-plan, /ccc-tdd, /ccc-verify
claude-code-best-practice	26K+	Reference architecture, Channels, Auto Mode
repomix	22.8K+	Pack codebases for AI (tree-sitter compression = 60% smaller)
oh-my-claudecode	17K+	HUD with worktree support, quota tracking, hyperlinks
Claude HUD	15K+	Real-time status display, offline cost tracking, git diffs
RTK	14.6K+	Token optimization (60-90% savings), 25 AWS subcommands
Compound Engineering	11.5K+	Knowledge compounding, mandatory code review enforcement
claude-skills	8.6K+	223+ skills, 23 agents, prompt A/B testing
claude-mem	46.7K+	Knowledge Agents, persistent cross-session memory
claude-code-ultimate-guide	2.7K+	219 templates, 271 quizzes, threat database
acpx	1.8K+	ACP protocol, Flows system, structured agent communication
claude-reflect	860+	Self-improving skills with reflection loops
Caliber	300+	Config scoring, drift detection
graphify	17.5K+	Any input → knowledge graph, clustered communities, HTML + JSON
UI/UX Pro Max	62K+	Design intelligence for professional UI/UX across platforms
claude-code-prompts	142+	Defensive prompt patterns, verification specialist
MengTo/Skills	100+	Web design + prompting — named-look design systems, video/HTML → prompt capture

---

Smart Orchestrator

The smart orchestrator scores each tool: capability match (50%) + popularity (15%) + recency (15%) + your preference (20%) — then picks the best one for each phase.

  PHASE          BEST TOOL              FALLBACK
  ──────────────────────────────────────────────
  ▸ Clarify      /ccc-office-hours          Spec flow
  ▸ Decide       /plan-ceo-review       Plan mode
  ▸ Plan         /ce:plan               Claude plan
  ▸ Execute      /ce:work               Dispatch
  ▸ Review       /ce:review (6+ agents) /simplify
  ▸ Test         /ccc-qa (real browser)     /ccc-verify
  ▸ Learn        Knowledge engine       Always on
  ▸ Ship         /ship                  git commit

CCC learns from every session. Knowledge compounds over time.

---

XRay + Makeover

Audit any project. Fix it automatically.

/ccc-xray                    # Scan → health score 0-100
/ccc-makeover                # Agent swarm applies top fixes

Dimension	Weight	What It Checks
Security	25%	CVEs, secrets, .env tracking
Testing	20%	Config, coverage, frameworks
Maintainability	20%	Complexity, linting, duplication
Dependencies	15%	Outdated, vulnerable
DevOps	10%	CI presence, quality gates
Documentation	10%	README, CLAUDE.md, inline docs

---

Quick Start — Pick Your Path

One question: How are you using Claude?

Path A: Desktop Plugin (Recommended)

The primary way to use CCC. Works in Claude Desktop and any Claude Code session.

/plugin marketplace add KevinZai/commander
/plugin install commander

CCC appears as a plugin you can invoke immediately. Say "start ccc" or "what should I build" to begin.

What you get: 71 plugin skills, 22 specialist agents, 23 lifecycle hooks (39 handlers), 2 credential-free bundled MCP servers + 16 opt-in via /ccc-connect. Free for now.

Marketplace: commander-hub at KevinZai/commander

---

Path B: Claude Code (Slash Commands)

You want skills + commands inside Claude Code sessions.

curl -fsSL https://raw.githubusercontent.com/KevinZai/commander/main/install-remote.sh | bash

Then in any Claude Code session:

/ccc

Full interactive menu appears. Same features, no separate CLI needed.

What you get: All 459+ skills and commands, no extra CLI binary required.

---

Your First 60 Seconds

1. ccc → Main menu (14 options)
2. Pick "Build something new" → Choose project type
3. Answer 3 multiple-choice questions → CCC generates a plan
4. CCC dispatches Claude with the right model, budget, and skills
5. Session ends → CCC extracts lessons → gets smarter next time

No configuration. No YAML. No API keys. The Intelligence Layer handles everything.

---

CLI Installation (Legacy)

You use claude in your terminal and prefer a standalone CLI.

Option 1 — One-liner install:

curl -fsSL https://raw.githubusercontent.com/KevinZai/commander/main/install-remote.sh | bash

Option 2 — npm:

npm install -g cc-commander

Then:

ccc

Arrow keys to navigate. Enter to select. That's it.

Update anytime:

ccc --update

Where files live:

• Source: ~/.cc-commander/ (auto-created by installer)
• Config: ~/.claude/ (skills, commands, hooks, CLAUDE.md)
• Binary: ccc → symlinked to source

Don't delete ~/.cc-commander/ — the ccc command needs it.

What you get: Full CLI with tmux split mode, daemon, theme switching, and the cockpit dashboard.

---

CLI Reference

Command	What It Does
ccc	Interactive mode (default — tmux tabbed)
ccc --stats	Sessions, streaks, level, cost
ccc --test	22-point self-test (verify install)
ccc --update	Pull latest + reinstall
ccc --repair	Reset corrupt state
ccc --simple	Menu-only, no tmux
ccc --dispatch "task"	Headless dispatch (for AI agents)
ccc --skills	Manage skill tiers (list, install, remove, tier)

Split Mode

The default mode. Tabbed tmux sessions. Each task gets its own window.

CCC menu runs in tab 0. Each dispatched task opens a new tmux window where Claude works with full output visible.

Key	Action
Ctrl+A n	Next tab
Ctrl+A p	Previous tab
Ctrl+A 0	Back to CCC menu
Ctrl+A q	Quit session
Mouse click	Switch tabs

Cancel Running Tasks

• During any build: Press Escape or q to kill the Claude process and return to menu
• During YOLO loop: touch ~/.claude/commander/yolo-stop to halt between cycles
• In split mode: Switch to the Claude tab and Ctrl+C

Daemon Mode

KAIROS-inspired persistent background agent. Monitors your project, processes queued tasks, and consolidates knowledge — all hands-free.

ccc --daemon                    # Start (runs in background)
ccc --queue "fix login bug"     # Add task to queue
ccc --queue-list                # Show pending tasks
ccc --daemon-stop               # Stop daemon

Feature	What It Does
Tick loop (5 min)	Checks queue, git status, dispatches work
Dream mode (1 hr)	Consolidates knowledge, detects error patterns
Task queue	Priority-based, file-backed, auto-dispatch
Budget cap	15-second limit per tick action

Customize: --interval 120 (2 min ticks) · --tick-budget 30 · --dream 30 (30 min dreams)

---

Use Inside Claude Code

No CLI needed. Type /ccc in any Claude Code session for the full interactive menu.

/ccc              → Main menu (15 options with sub-menus)
/ccc xray         → Project health scan
/ccc makeover     → Auto-apply top fixes
/ccc refresh      → Update your CLAUDE.md from latest template
/ccc domains      → Browse 11 CCC domains
/ccc skills       → Browse 502+ skills
/ccc grill        → 7-question Socratic planning probe
/ccc infra        → Infrastructure sub-menu (Fleet, Synapse, Cost, AO, CloudCLI, Paperclip)
/ccc detect       → Probe all services and CLIs

Same choices, same sub-menus, same actions as the full CLI. Cancel anytime with "back" or Escape. Also works in Claude Desktop Cowork and VS Code / Cursor.

---

Agent-Friendly API

CCC is built to be controlled by AI agents — OpenClaw, Claude Code, or any orchestrator.

Command	Output	Purpose
ccc --dispatch "task" --json	JSON	Run task headlessly
ccc --list-skills --json	JSON	All 459+ skills
ccc --list-sessions --json	JSON	Session history
ccc --status	JSON	Health check
ccc --template	text	Latest CLAUDE.md template

Override flags: --model opus · --max-turns 50 · --budget 5 · --cwd /path

# OpenClaw agent dispatches a build
result=$(ccc --dispatch "Build auth with JWT" --json --model opus --budget 5)

# Claude Code agent checks available skills
ccc --list-skills --json | jq '.[] | select(.name | contains("auth"))'

---

Security

• No command injection — all shell calls use execFileSync with array arguments
• Auto-mode by default — dispatches use --permission-mode auto, not --dangerously-skip-permissions
• YOLO only skips — only YOLO/night mode uses skip-permissions (intentional, documented)
• Zero npm vulnerabilities — verified on every CI run
• PII scanning — CI blocks commits containing personal data patterns
• Error containment — every error returns to menu with error ID, never raw stack traces
• Pre-publish checks — npm run prepublishOnly runs all tests + lint before publish

---

Agent Bible

BIBLE-AGENT.md — 268-line machine-readable reference. Any AI agent reads this and can manage CCC immediately.

# Tell any agent:
"Read docs/BIBLE-AGENT.md from the cc-commander repo, then use the CLI API to manage this project."

Covers: CLI API, dispatch patterns, JSON schemas, skill catalog, level/model defaults, integration points.

---

The Kevin Z Method

7 rules from 200+ articles. 14 months of production.

1. Plan before coding
2. Context is milk — keep it fresh
3. Verify, don't trust
4. Subagents = fresh context
5. CLAUDE.md is an investment
6. Boring solutions win
7. Operationalize every fix

Full methodology: BIBLE.md — 2000+ lines, 7 chapters, appendices.

For AI agents: BIBLE-AGENT.md — 268-line machine-readable version.

---

10 Themes

Cyberpunk, Fire, Ocean, Aurora, Sunset, Monochrome, Rainbow, Dracula, Graffiti, Futuristic. Live preview as you navigate. Switch anytime in Settings or type /ccc theme.

---

vs aider.chat — positioning

Both tools make you faster. They solve different problems.

Dimension	aider	CC Commander
Surface	Terminal / Git	Claude Code Desktop + CLI + IDE
LLM	Any (OpenAI, Anthropic, local)	Claude-native (Sonnet / Opus / Haiku)
Core pattern	Pair programmer — you drive, aider edits	Guided PM — CCC plans, routes, dispatches
Primitives	Diff-based file editing + Git	Claude Agent SDK: skills, agents, hooks, MCP
Agent model	Single model, single context	22 specialist sub-agent personas (brain/hands)
Git	Deep — commit messages, repo mapping	Lifecycle hooks + /ccc-ship pre-flight
Context	Repo map (tree-sitter, 15k tokens)	Session compounding + knowledge base
Cost	API pay-per-token	Free for now (affiliate + consulting model)
Install	pip install aider-chat	Desktop plugin marketplace or curl | bash
Best at	Solo rapid edits, refactors, small files	Multi-phase features, reviews, fleet orchestration

Aider is your pair programmer. CC Commander is your PM. They're complementary — use aider for solo rapid edits, CCC for multi-phase features that need structure.

---

Who Built This

CCC is built by Kevin Z (@kzic) — a non-technical entrepreneur and CEO of MyWiFi Networks (20+ years in tech, never wrote production code). He wanted to leverage Claude Code to its max capabilities without chasing plugins, reading changelogs, or memorizing commands. So he scanned every Claude Code article, plugin, and skill on the internet — 200+ sources — and distilled it into one self-learning AI project manager. Every feature was built with Claude Code itself. If a non-coder can build a 450+ skill toolkit with AI, imagine what you can build.

---

Acknowledgments

CCC aggregates 19 open-source packages. Full credits: ACKNOWLEDGMENTS.md

45+ ecosystem repos tracked: ECOSYSTEM.md

---

Share

If CC Commander helped you ship faster, share it:

License

MIT License for the full project. The Intelligence Layer (4 files) has an additional Commons Clause — free to use, not to sell.

All 19 vendor packages are permissive open-source (MIT, Apache-2.0, and CC-BY-SA-4.0).

Note: GitHub may show "Unknown" in the sidebar because some vendor submodules don't ship a LICENSE file in their repo root. Each vendor retains its upstream license — check the linked vendor repositories for specifics.

---

Contributing

skills/your-skill/SKILL.md        # Add a skill
commands/your-command.md           # Add a command
hooks/your-hook.js                 # Add a hook
commander/adventures/X.json        # Add a flow

---