Getting Started

Get CCCC running in 10 minutes.

Choose Your Approach

CCCC offers two ways to get started:

Web UI Quick Start

Recommended for most users

• Visual interface for managing agents
• Point-and-click configuration
• Real-time terminal view
• Mobile-friendly

CLI Quick Start

For terminal enthusiasts

• Full control via command line
• Scriptable and automatable
• Great for CI/CD integration
• Power user features

Docker Deployment

For servers and teams

• One-command deployment
• Pre-installed AI agent CLIs
• Persistent data with volumes
• Docker Compose and K8s ready

Prerequisites

Both approaches require:

• Python 3.11+ installed
• At least one AI agent CLI:
  • Claude Code (recommended)
  • Codex CLI
  • GitHub Copilot CLI
  • Cursor CLI
  • Devin CLI
  • Kiro CLI
  • Kilo Code CLI
  • Antigravity CLI
  • Kimi CLI
• Or a ChatGPT account with remote MCP connector support for the ChatGPT Web Model runtime
• Or a custom runtime command if you wire MCP manually

The MCP JavaScript code mode (cccc_code_exec / cccc_code_wait) additionally requires Node.js on the CCCC host. The standalone Rust build does not invoke a Python backend.

The ChatGPT Web Model also needs a system Google Chrome or Microsoft Edge browser. On native Linux, install Xvfb so CCCC can keep projected browser windows off the host desktop; x11vnc is optional and enables the VNC viewer instead of the built-in CDP screencast fallback:

# Debian / Ubuntu
sudo apt install xvfb x11vnc

# Fedora
sudo dnf install xorg-x11-server-Xvfb x11vnc

# Arch Linux
sudo pacman -S xorg-server-xvfb x11vnc

Run cccc doctor to verify these dependencies. CCCC does not install OS packages automatically.

Installation

Upgrading from older versions

If you have an older version of cccc-pair installed (e.g., 0.3.x), you must uninstall it first:

# For pipx users
pipx uninstall cccc-pair

# For pip users
pip uninstall cccc-pair

# Remove any leftover binaries if needed
rm -f ~/.local/bin/cccc ~/.local/bin/ccccd

Version 0.4.x Breaking Changes

Version 0.4.x has a completely different command structure from 0.3.x. The old init, run, bridge commands are replaced with attach, daemon, mcp, etc.

From PyPI (recommended)

python -m pip install -U cccc-pair

PyPI is the stable CCCC product distribution, with Python as its stable and recommended implementation. Supported platform wheels also contain a private, version-matched experimental Rust implementation for opt-in performance evaluation with cccc rust. Complete feature and integration parity is not promised; Rust is a bounded evaluation surface for measured native benefits. Use cccc python for reliability-critical workflows. Other platforms receive the universal Python wheel.

Experimental standalone Rust preview

Rust-only preview

Use this optional channel only to evaluate deployment without Python. It has no Python fallback or implementation switching and is not yet the recommended replacement for the complete pip product.

macOS or Linux:

curl -fsSL https://chesterra.github.io/cccc/install.sh | sh

Windows CMD or PowerShell:

powershell.exe -NoProfile -ExecutionPolicy Bypass -Command "[Net.ServicePointManager]::SecurityProtocol = [Net.ServicePointManager]::SecurityProtocol -bor [Net.SecurityProtocolType]::Tls12; Invoke-RestMethod 'https://chesterra.github.io/cccc/install.ps1' | Invoke-Expression"

The installer downloads a checksum-verified GitHub Release binary. It does not require a Rust or Python toolchain and refuses to overwrite an existing cccc command it does not own. Uninstall that command deliberately or choose another CCCC_INSTALL_DIR first. The hosted installer stays pinned to the newest published release with a complete asset set. You can select any fully published version explicitly with CCCC_VERSION:

Commands in other directories are never removed. The default installer puts its directory first in the user PATH and reports every remaining cccc command. Open a new terminal after installation and run cccc doctor to verify that PATH resolves to the executable you just installed.

curl -fsSL https://chesterra.github.io/cccc/install.sh | CCCC_VERSION=0.4.34 sh

$env:CCCC_VERSION = "0.4.34"
powershell.exe -NoProfile -ExecutionPolicy Bypass -Command "[Net.ServicePointManager]::SecurityProtocol = [Net.ServicePointManager]::SecurityProtocol -bor [Net.SecurityProtocolType]::Tls12; Invoke-RestMethod 'https://chesterra.github.io/cccc/install.ps1' | Invoke-Expression"
Remove-Item Env:CCCC_VERSION

From TestPyPI (for explicit RC testing)

python -m pip install -U --pre \
  --index-url https://test.pypi.org/simple \
  --extra-index-url https://pypi.org/simple \
  cccc-pair

From Source

git clone https://github.com/ChesterRa/cccc
cd cccc
pip install -e .

Verify Installation

cccc status
cccc doctor

status shows the selected, running, and available product implementations. Python is the stable initial default while no implementation choice has been saved. On a supported platform wheel, use cccc rust to opt into the bundled experimental Rust implementation or cccc python to return to stable Python; later bare invocations follow that persisted choice. doctor checks Python, agent runtimes, system configuration, the invoked CCCC executable, PATH resolution, and duplicate cccc commands.

Next Steps

• Web UI Quick Start - Get started with the visual interface
• CLI Quick Start - Get started with the command line
• Docker Deployment - Deploy CCCC in a Docker container
• SDK Overview - Integrate CCCC into external apps/services
• Use Cases - Learn high-ROI real-world patterns
• Operations Runbook - Run CCCC with operator-grade reliability
• Positioning - Decide where CCCC should sit in your stack