Craig-O-Code Logo

Craig-O-Code

Agent-Powered CLI for Claude Code

© 2026 VibeCaaS.com | NeuralQuantum.ai LLC

Installation •
Usage •
Agents •
Commands

---

NeuralQuantum.ai Agent-Powered CLI - A command-line tool that wraps Claude Code with 68+ specialized AI agents for automated development workflows.

> ⚠️ Internal Version: This build includes an embedded OAuth token for internal VibeCaaS/NeuralQuantum.ai team use only. Do not distribute publicly.

---

Features

- Auto-Orchestration: Automatically analyzes, enhances, and routes your requests to the right agents
- Chain-of-Verification (CoV): High-accuracy, bias-resistant verification workflow for critical decisions
- 68+ Specialized Agents: AI, Quantum, iOS, Web, DevOps, Security, and more
- Metacognition Layer: Self-aware reasoning with critique, regulation, and learning
- Fast Dispatch: Quick routing for simple tasks
- Project Integration: Initialize any project with full agent support

---

Pricing

| Tier | Price | Agents | Features |
|------|-------|--------|----------|
| Free | $0 | 68 | Basic orchestration, CoV workflow |
| Pro | $29/mo | 100+ | Advanced orchestration, priority support, pro agents |
| Enterprise | Custom | Unlimited | Custom agents, dedicated support, SLA |

$3

``bash

`Set your license key`


coc license set COC-PRO-XXXXXXXX
Check license status

coc license info


Get your Pro license at vibecaas.com/craig-o-code/pro
---
Prerequisites

- Claude Code - Install from claude.ai/code`bash npm install -g @anthropic-ai/claude-code claude auth login`

- macOS, Linux, or WSL (Windows Subsystem for Linux)

---

`Installation`

`$3`

The easiest way to install Craig-O-Code:

`bash npm install -g craig-o-code`

That's it! The package will: - Install thecoc, craig, and craig-o-codeCLI commands globally - Set up 68+ agents in~/.cursor/agents/- Create global cursor rules

Verify installation:`bash coc version coc list`

`$3`

`bash

`Clone the repository`


git clone https://github.com/ttracx/craig-o-code.git
cd craig-o-code
Run installer

chmod +x scripts/install.sh
./scripts/install.sh
Reload your shell

source ~/.zshrc  # or ~/.bashrc

$3

`bash curl -fsSL https://raw.githubusercontent.com/ttracx/craig-o-code/master/install.sh | bash`

This one-liner will: - Install Claude Code if not present - Set up the Anthropic API key - Install Craig-O-Code globally - Install 68+ agents

`$3`

Update Craig-O-Code and all agents to the latest version:

`bash curl -fsSL https://raw.githubusercontent.com/ttracx/craig-o-code/master/update.sh | bash`

`$3`

If you've already installed via npm and want to update agents:

`bash npm run install-agents`

---

`Usage`

`$3`

Just describe what you want - Craig-O-Code figures out the rest:

`bash coc go "add user authentication with JWT" coc go "build a REST API for products" coc go "fix the memory leak in the worker"`

`$3`

For simple, single-domain tasks:

`bash coc quick "add a loading spinner" coc quick "fix the button hover state"`

`$3`

For critical decisions requiring verified, bias-resistant answers:

`bash coc verify "What's the best approach for database migrations?" coc deepverify "Should we use microservices or monolith?" coc quickverify "Is Redis good for session storage?"`

`$3`

When you know exactly which specialist you need:

`bash coc agent swiftui-architect "design a settings view" coc agent quantum-ml-researcher "implement VQE algorithm" coc agent rag-specialist "design hybrid search"`

`$3`

Even faster - just use craig:

`bash craig "add dark mode support" craig "optimize database queries" craig "build a dashboard component"`

---

`Commands Reference`

`$3`

| Command | Description | |---------|-------------| |coc go | Full auto-orchestration pipeline | |coc do | Alias for go| |coc quick | Fast routing for simple tasks | |craig | Shortcut for coc go |

`$3`

| Command | Description | |---------|-------------| |coc agent | Invoke specific agent | |coc list| List all available agents | |coc teams| Show agent teams | |coc info | Show agent details |

`$3`

| Command | Description | |---------|-------------| |coc critique | MCL critique of code/plan | |coc review | Quick code review | |coc learn | After-action review |

`$3`

| Command | Description | |---------|-------------| |coc verify | Full CoV verification workflow | |coc cov | Alias for verify| |coc quickverify | Quick verification (3 workers) | |coc deepverify | Deep verification (5+ workers) |

`$3`

| Command | Description | |---------|-------------| |coc init| Initialize project with agents | |coc update| Update agents from global | |coc orchestrate | Full project orchestration |

`$3`

| Command | Description | |---------|-------------| |coc shell| Interactive Claude Code | |coc config| Show configuration | |coc version| Show version | |coc help | Full documentation |

---

`Agent Teams`

`$3`


Automated prompt routing and enhancement
-

auto-orchestrator

 - Master pipeline controller
-

prompt-analyzer

 - Intent extraction
-

prompt-enhancer

 - Request enrichment
-

intent-router

 - Agent selection
-

smart-dispatcher

 - Fast routing
$3

Self-aware reasoning and learning
-

mcl-core

 - Central controller
-

mcl-critic

 - Output evaluation
-

mcl-regulator

 - Impulse control
-

mcl-learner

 - Learning from outcomes
-

agent-creator

 - Generate new agents
$3

High-accuracy, bias-resistant verification
-

cov-orchestrator

 - CoV workflow controller
-

domain-expert-worker

 - Domain-specific factual validation
-

counterexample-hunter-worker

 - Find edge cases, failure modes
-

comparative-analyst-worker

 - Compare alternatives, tradeoffs
-

risk-safety-reviewer-worker

 - Identify risks, policy concerns
-

clarity-editor-worker

 - Improve clarity without changing meaning
$3

-

llm-integration-architect

rag-specialist

prompt-engineer

ai-agent-architect

fine-tuning-specialist

ml-ops-engineer


$3

-

quantum-circuit-designer

quantum-algorithm-developer

quantum-ml-researcher

quantum-error-specialist

quantum-simulation-expert


$3

-

swiftui-architect

ios-performance-engineer

swift-concurrency-expert

apple-platform-integrator

ios-testing-specialist


$3

-

frontend-architect

backend-architect

fullstack-developer


$3

-

cloud-architect

ci-cd-engineer

kubernetes-specialist


$3

-

security-architect

appsec-engineer


---
Examples
$3

`bash

`Full-stack feature`


coc go "build a user profile page with avatar upload"
API development

coc go "create REST endpoints for order management"
iOS feature

coc agent swiftui-architect "design a settings screen with dark mode toggle"

$3

`bash

`Bug fix`


coc quick "fix the null pointer in UserService"
Performance

coc go "optimize the slow database queries"
Security

coc agent appsec-engineer "audit authentication module"

$3

`bash

`Review`


coc critique src/components/
Refactor

coc go "refactor auth module for better testability"
Documentation

coc agent doc-generator "document the API endpoints"

$3

`bash

`Large projects`


coc orchestrate "build an e-commerce platform with React and Node"
coc orchestrate "migrate monolith to microservices"

$3

`bash

`Architectural decisions`


coc verify "What's the best database for our high-write workload?"
coc deepverify "Should we use event sourcing or CRUD?"
Technical recommendations

coc verify "Is our caching strategy optimal for read-heavy traffic?"
coc quickverify "Should we use WebSockets or SSE for real-time updates?"


---
Configuration
Craig-O-Code stores configuration in:

- CLI tools: ~/.local/bin/- Agents:~/.cursor/agents/- Rules:~/.cursorrules

`$3`

Create ~/.coc-config for custom settings:

`yaml default_mode: balanced verbose: false auto_include: - code-reviewer - test-generator`

---

`Uninstall`

`$3`

`bash npm uninstall -g craig-o-code`

`$3`

`bash ./scripts/uninstall.sh`

`$3`

`bash rm ~/.local/bin/coc ~/.local/bin/craig ~/.local/bin/craig-o-code rm -rf ~/.cursor/agents rm ~/.cursorrules rm ~/.craig-o-code-intro.sh`

---

`Troubleshooting`

`$3`

Install Claude Code:`bash npm install -g @anthropic-ai/claude-code claude auth login`

`$3`

If installed via npm, ensure your npm global bin directory is in PATH:`bash npm config get prefix # Shows npm prefix export PATH="$(npm config get prefix)/bin:$PATH"`

If installed from source, ensure ~/.local/binis in your PATH:`bash export PATH="$HOME/.local/bin:$PATH"`

Add to ~/.zshrc or ~/.bashrc for persistence.

`$3`

Reinstall agents via npm:`bash npm run install-agents`

Or from source:`bash ./scripts/install.sh --agents-only`

`$3`

If you encounter EACCES errors:`bash

`Option 1: Use a node version manager (recommended)`


nvm install --lts
nvm use --lts
Option 2: Fix npm permissions

mkdir ~/.npm-global
npm config set prefix '~/.npm-global'
export PATH=~/.npm-global/bin:$PATH


---
Project Structure

`craig-o-code/ ├── agents/ # 68+ specialized AI agent definitions (including CoV team) │ ├── metacognition/ # MCL (Metacognition Layer) agents │ ├── teams/ # Organized agent teams │ └── skills/ # Reusable agent skills ├── bin/ # Executable CLI tools │ ├── coc # Main CLI entry point │ ├── craig # Shortcut alias │ └── auto-route.sh # Smart routing logic ├── docs/ # Documentation │ ├── ALIASES.md # Shell alias reference │ ├── COV-QUICKSTART.md # CoV update & usage guide │ ├── FUNCTIONGEMMA_CONFIG.md # Local model configuration │ ├── DISTRIBUTION-READY.md # Distribution guide │ └── adr/ # Architecture decision records ├── macos-installer/ # macOS installer build system │ ├── build-all.sh # Complete build pipeline │ ├── build-dist.sh # Create DMG and ZIP │ └── build-pkg.sh # Build installer package ├── resources/ # Images and assets │ └── Craig-O-Code.png # Project logo ├── scripts/ # Utility scripts │ ├── install.sh # Installation script (bash) │ ├── npm-postinstall.js # npm postinstall script (Node.js) │ ├── uninstall.sh # Uninstallation script │ └── QUICK_RELOAD.sh # Alias reload helper ├── CHANGELOG.md # Version history ├── README.md # This file ├── VERSION # Current version number └── package.json # NPM package definition``

---

License

MIT License - NeuralQuantum.ai

---

Links

- VibeCaaS.com: https://vibecaas.com
- NeuralQuantum.ai: https://NeuralQuantum.ai
- Claude Code: https://claude.ai/code
- Issues: https://github.com/ttracx/craig-o-code/issues

---

Craig-O-Code Logo

Craig-O-Code

Agent-Powered CLI for Claude Code

© 2026 VibeCaaS.com | NeuralQuantum.ai LLC

Installation •
Usage •
Agents •
Commands

---

NeuralQuantum.ai Agent-Powered CLI - A command-line tool that wraps Claude Code with 68+ specialized AI agents for automated development workflows.

> ⚠️ Internal Version: This build includes an embedded OAuth token for internal VibeCaaS/NeuralQuantum.ai team use only. Do not distribute publicly.

---

Features

---

Pricing

$3

``bash

`Set your license key`


coc license set COC-PRO-XXXXXXXX
Check license status

coc license info


Get your Pro license at vibecaas.com/craig-o-code/pro
---
Prerequisites

- Claude Code - Install from claude.ai/code`bash npm install -g @anthropic-ai/claude-code claude auth login`

- macOS, Linux, or WSL (Windows Subsystem for Linux)

---

`Installation`

`$3`

The easiest way to install Craig-O-Code:

`bash npm install -g craig-o-code`

That's it! The package will: - Install thecoc, craig, and craig-o-codeCLI commands globally - Set up 68+ agents in~/.cursor/agents/- Create global cursor rules

Verify installation:`bash coc version coc list`

`$3`

`bash

`Clone the repository`


git clone https://github.com/ttracx/craig-o-code.git
cd craig-o-code
Run installer

chmod +x scripts/install.sh
./scripts/install.sh
Reload your shell

source ~/.zshrc  # or ~/.bashrc

$3

`bash curl -fsSL https://raw.githubusercontent.com/ttracx/craig-o-code/master/install.sh | bash`

This one-liner will: - Install Claude Code if not present - Set up the Anthropic API key - Install Craig-O-Code globally - Install 68+ agents

`$3`

Update Craig-O-Code and all agents to the latest version:

`bash curl -fsSL https://raw.githubusercontent.com/ttracx/craig-o-code/master/update.sh | bash`

`$3`

If you've already installed via npm and want to update agents:

`bash npm run install-agents`

---

`Usage`

`$3`

Just describe what you want - Craig-O-Code figures out the rest:

`bash coc go "add user authentication with JWT" coc go "build a REST API for products" coc go "fix the memory leak in the worker"`

`$3`

For simple, single-domain tasks:

`bash coc quick "add a loading spinner" coc quick "fix the button hover state"`

`$3`

For critical decisions requiring verified, bias-resistant answers:

`bash coc verify "What's the best approach for database migrations?" coc deepverify "Should we use microservices or monolith?" coc quickverify "Is Redis good for session storage?"`

`$3`

When you know exactly which specialist you need:

`bash coc agent swiftui-architect "design a settings view" coc agent quantum-ml-researcher "implement VQE algorithm" coc agent rag-specialist "design hybrid search"`

`$3`

Even faster - just use craig:

`bash craig "add dark mode support" craig "optimize database queries" craig "build a dashboard component"`

---

`Commands Reference`

`$3`

| Command | Description | |---------|-------------| |coc critique | MCL critique of code/plan | |coc review | Quick code review | |coc learn | After-action review |

`$3`

| Command | Description | |---------|-------------| |coc init| Initialize project with agents | |coc update| Update agents from global | |coc orchestrate | Full project orchestration |

`$3`

| Command | Description | |---------|-------------| |coc shell| Interactive Claude Code | |coc config| Show configuration | |coc version| Show version | |coc help | Full documentation |

---

`Agent Teams`

`$3`


Automated prompt routing and enhancement
-

auto-orchestrator

 - Master pipeline controller
-

prompt-analyzer

 - Intent extraction
-

prompt-enhancer

 - Request enrichment
-

intent-router

 - Agent selection
-

smart-dispatcher

 - Fast routing
$3

Self-aware reasoning and learning
-

mcl-core

 - Central controller
-

mcl-critic

 - Output evaluation
-

mcl-regulator

 - Impulse control
-

mcl-learner

 - Learning from outcomes
-

agent-creator

 - Generate new agents
$3

High-accuracy, bias-resistant verification
-

cov-orchestrator

 - CoV workflow controller
-

domain-expert-worker

 - Domain-specific factual validation
-

counterexample-hunter-worker

 - Find edge cases, failure modes
-

comparative-analyst-worker

 - Compare alternatives, tradeoffs
-

risk-safety-reviewer-worker

 - Identify risks, policy concerns
-

clarity-editor-worker

 - Improve clarity without changing meaning
$3

-

llm-integration-architect

rag-specialist

prompt-engineer

ai-agent-architect

fine-tuning-specialist

ml-ops-engineer


$3

-

quantum-circuit-designer

quantum-algorithm-developer

quantum-ml-researcher

quantum-error-specialist

quantum-simulation-expert


$3

-

swiftui-architect

ios-performance-engineer

swift-concurrency-expert

apple-platform-integrator

ios-testing-specialist


$3

-

frontend-architect

backend-architect

fullstack-developer


$3

-

cloud-architect

ci-cd-engineer

kubernetes-specialist


$3

-

security-architect

appsec-engineer


---
Examples
$3

`bash

`Full-stack feature`


coc go "build a user profile page with avatar upload"
API development

coc go "create REST endpoints for order management"
iOS feature

coc agent swiftui-architect "design a settings screen with dark mode toggle"

$3

`bash

`Bug fix`


coc quick "fix the null pointer in UserService"
Performance

coc go "optimize the slow database queries"
Security

coc agent appsec-engineer "audit authentication module"

$3

`bash

`Review`


coc critique src/components/
Refactor

coc go "refactor auth module for better testability"
Documentation

coc agent doc-generator "document the API endpoints"

$3

`bash

`Large projects`


coc orchestrate "build an e-commerce platform with React and Node"
coc orchestrate "migrate monolith to microservices"

$3

`bash

`Architectural decisions`


coc verify "What's the best database for our high-write workload?"
coc deepverify "Should we use event sourcing or CRUD?"
Technical recommendations

coc verify "Is our caching strategy optimal for read-heavy traffic?"
coc quickverify "Should we use WebSockets or SSE for real-time updates?"


---
Configuration
Craig-O-Code stores configuration in:

- CLI tools: ~/.local/bin/- Agents:~/.cursor/agents/- Rules:~/.cursorrules

`$3`

Create ~/.coc-config for custom settings:

`yaml default_mode: balanced verbose: false auto_include: - code-reviewer - test-generator`

---

`Uninstall`

`$3`

`bash npm uninstall -g craig-o-code`

`$3`

`bash ./scripts/uninstall.sh`

`$3`

`bash rm ~/.local/bin/coc ~/.local/bin/craig ~/.local/bin/craig-o-code rm -rf ~/.cursor/agents rm ~/.cursorrules rm ~/.craig-o-code-intro.sh`

---

`Troubleshooting`

`$3`

Install Claude Code:`bash npm install -g @anthropic-ai/claude-code claude auth login`

`$3`

If installed via npm, ensure your npm global bin directory is in PATH:`bash npm config get prefix # Shows npm prefix export PATH="$(npm config get prefix)/bin:$PATH"`

If installed from source, ensure ~/.local/binis in your PATH:`bash export PATH="$HOME/.local/bin:$PATH"`

Add to ~/.zshrc or ~/.bashrc for persistence.

`$3`

Reinstall agents via npm:`bash npm run install-agents`

Or from source:`bash ./scripts/install.sh --agents-only`

`$3`

If you encounter EACCES errors:`bash

`Option 1: Use a node version manager (recommended)`


nvm install --lts
nvm use --lts
Option 2: Fix npm permissions

mkdir ~/.npm-global
npm config set prefix '~/.npm-global'
export PATH=~/.npm-global/bin:$PATH


---
Project Structure

---

License

MIT License - NeuralQuantum.ai

---

Links

- VibeCaaS.com: https://vibecaas.com
- NeuralQuantum.ai: https://NeuralQuantum.ai
- Claude Code: https://claude.ai/code
- Issues: https://github.com/ttracx/craig-o-code/issues

---