Quick Install

Two commands to get everything running.

Step 1: Install

Run this single command to install all OrcAI CLI tools (Claude Code, Codex CLI, Gemini CLI, ccswitch):

Step 2: Setup

Configure your API key with ccswitch. This sets up all tools at once:

What Gets Installed

System Requirements

ccswitch

The unified configuration tool for all OrcAI CLI tools.

ccswitch is a command-line tool that manages API keys and endpoint configuration for Claude Code, Codex CLI, and Gemini CLI. Instead of manually setting environment variables, use ccswitch to configure everything at once.

Commands

ccswitch setup

Configure your API key and set up all CLI tools to use OrcAI gateway:

# Initial setup with your API key
$ ccswitch setup YOUR_API_KEY

# This automatically configures:
# - ANTHROPIC_API_KEY + ANTHROPIC_BASE_URL (for Claude Code)
# - OPENAI_API_KEY + OPENAI_BASE_URL (for Codex CLI)
# - GEMINI_API_KEY (for Gemini CLI)

ccswitch status

Check the current configuration status of all tools:

$ ccswitch status

# Output:
Claude Code  ✓ configured (api.orcai.cc)
Codex CLI    ✓ configured (api.orcai.cc)
Gemini CLI   ✓ configured (api.orcai.cc)
API Key      sk-****...****

ccswitch use

Switch between CLI tools or launch a specific tool:

# Launch Claude Code
$ ccswitch use claude

# Launch Codex CLI
$ ccswitch use codex

# Launch Gemini CLI
$ ccswitch use gemini

ccswitch help

Show help and available commands:

$ ccswitch help

# Available commands:
#   setup <api-key>   Configure API key for all tools
#   status            Show current configuration
#   use <tool>        Launch a specific CLI tool
#   help              Show this help message

Command Summary

Claude Code

Anthropic's agentic coding tool that lives in your terminal.

Claude Code understands your entire codebase, executes routine tasks, explains complex code, and handles git workflows through natural language commands.

Launch

# Via ccswitch
$ ccswitch use claude

# Or directly (after ccswitch setup)
$ claude

Key Features

• Multi-file editing — Search and modify files across your entire codebase
• Git integration — Read issues, write code, run tests, submit PRs
• Agentic search — Automatically finds relevant context without manual selection
• MCP support — Connect external tools via Model Context Protocol
• Slash commands — Built-in commands like /commit, /review, and custom plugins

Usage Examples

# Interactive session
$ claude

# Direct task
$ claude "explain this function"

# Fix a bug
$ claude "fix the login bug in auth.js"

# Create a commit
$ claude "/commit"

# Update Claude Code
$ claude update

Supported Models

Codex CLI

OpenAI's terminal-based coding agent built in Rust.

Codex CLI reads, changes, and runs code on your machine with sandboxed execution for security. Built primarily in Rust for maximum performance.

Launch

# Via ccswitch
$ ccswitch use codex

# Or directly (after ccswitch setup)
$ codex

Key Features

• Built in Rust — 96% Rust codebase for maximum performance
• Sandboxed execution — Secure code execution with Landlock/seccomp on Linux
• Editor integration — Works with VS Code, Cursor, and Windsurf
• Desktop app — Launch with codex app for a graphical interface
• Web search — Built-in web search for gathering context

Usage Examples

# Interactive mode
$ codex

# Direct task
$ codex "Build a React dashboard component"

# Review code
$ codex "/review"

# Desktop app mode
$ codex app

Gemini CLI

Google's open-source AI agent with 1M token context window.

Gemini CLI is powered by Gemini models with a massive 1M token context window. It includes a free tier with generous limits and supports Google Search grounding and MCP servers.

Launch

# Via ccswitch
$ ccswitch use gemini

# Or directly (after ccswitch setup)
$ gemini

Key Features

• 1M token context — Largest context window of any CLI tool
• Free tier — 60 requests/min, 1,000 requests/day at no cost
• Open source — Apache 2.0 license, fully transparent
• Google Search — Real-time web search grounding
• Checkpointing — Save and resume conversation sessions
• Non-interactive mode — Scriptable for automation pipelines

Free Tier Limits

OpenClaw

Multi-channel AI bot framework for 13+ messaging platforms.

OpenClaw is a personal AI assistant framework that runs on your own devices and connects to Telegram, Discord, Slack, WhatsApp, and more. It supports voice, browser control, and a powerful skill system.

Install

# Run onboarding
$ openclaw onboard --install-daemon

Telegram Setup

1. Create a bot via @BotFather on Telegram (/newbot)
2. Configure the bot token in OpenClaw config:

{
  channels: {
    telegram: {
      enabled: true,
      botToken: "YOUR_BOT_TOKEN",
      dmPolicy: "pairing"
    }
  }
}

# Start gateway and approve pairing
$ openclaw gateway
$ openclaw pairing list telegram
$ openclaw pairing approve telegram <CODE>

Supported Platforms

API Overview

OpenAI-compatible REST API for accessing 50+ AI models.

Base URL

https://api.orcai.cc/v1

Quick Example

curl https://api.orcai.cc/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "messages": [{"role": "user", "content": "Hello!"}],
    "max_tokens": 1024
  }'

from openai import OpenAI

client = OpenAI(
    api_key="your-orcai-api-key",
    base_url="https://api.orcai.cc/v1"
)

response = client.chat.completions.create(
    model="claude-sonnet-4-6",
    messages=[{"role": "user", "content": "Hello!"}],
    max_tokens=1024
)
print(response.choices[0].message.content)

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'your-orcai-api-key',
  baseURL: 'https://api.orcai.cc/v1'
});

const response = await client.chat.completions.create({
  model: 'claude-sonnet-4-6',
  messages: [{ role: 'user', content: 'Hello!' }],
  max_tokens: 1024
});
console.log(response.choices[0].message.content);

Endpoints

Authentication

Secure your API requests with bearer token authentication.

API Key

All API requests require a valid API key in the Authorization header:

Authorization: Bearer YOUR_API_KEY

Getting Your API Key

1. Log in to OrcAI Portal
2. Navigate to the API Keys section
3. Click Create New Key
4. Copy the key immediately - it won't be shown again

Rate Limits

Models

50+ AI models from leading providers, accessible through one API.

Anthropic Models

OpenAI Models

Google Models