Skip to main content

Introduction

Bring Your Own AI(BYOA) team in Slack/IDE(MCP) with your existing subscriptions

Transform Claude, Gemini, Codex and Copilot into a collaborative development team. No extra costs—just your existing AI subscriptions working together.

Why CrewX?​

Autopilot, Not Copilot - AI That Works While You Sleep​

CrewX and Claude Code work better together:

Claude CodeCrewX
🧑‍✈️ Copilot - You drive, AI assists🤖 Autopilot - AI team works autonomously
Sequential, hands-on developmentParallel execution across multiple tasks
Individual developer focusTeam collaboration in Slack
Perfect for prototyping & experimentationPerfect for scaling & automation

The Ultimate AI Development Workflow:

1. Claude Code → Prototype & experiment with ideas
2. CrewX → Multiple agents build features in parallel
3. Claude Code → Final review & quality polish

Pro Tip: Claude Code as Engineering Manager

Use Claude Code to orchestrate your CrewX team like a development manager:

# .claude/commands/crewx-dev-lead.md
You are the CrewX Development Lead managing a team of AI agents.

## Your Role
✅ Plan releases and prioritize work
✅ Delegate tasks to specialized agents
✅ Monitor progress and coordinate
✅ Make decisions and report status

❌ Never write code directly (delegate to @dev agents)
❌ Never run tests directly (delegate to @qa agents)
❌ Never deploy directly (delegate to @release agents)

## Team Structure (via crewx.yaml)
- @crewx_claude_dev - Senior developer (coding, debugging)
- @crewx_qa_lead - QA team lead (test planning)
- @crewx_qa_tester - Test execution specialist
- @crewx_release_manager - Release & deployment expert

## Example Workflow
# 1. Fix bug in parallel
crewx execute "@crewx_claude_dev Fix bug #123" "@crewx_qa_lead Create test plan"

# 2. Run tests after fix
crewx execute "@crewx_qa_lead Test bug #123"

# 3. Release when ready
crewx execute "@crewx_release_manager Create release 1.0.0-rc.0"

Real-world example: The CrewX project itself uses this pattern—Claude Code acts as dev lead, delegating to specialized CrewX agents for implementation, testing, and releases.

Claude Code stays strategic while CrewX agents handle tactical execution—like a CTO directing engineering teams.

CrewX doesn't replace your favorite AI coding assistant—it amplifies it by handling the heavy lifting while you focus on creative decisions.

Slack Team Collaboration - Your AI Team in Slack​

Bring AI agents directly into your team's workspace:

  • Team-wide AI access - Everyone benefits from AI expertise in Slack channels
  • Thread-based context - Maintains conversation history automatically
  • Multi-agent collaboration - @claude, @gemini, @copilot work together in real-time
  • Natural integration - Works like chatting with team members
  • Shared knowledge - Team learns from AI interactions, not isolated sessions

Remote Agents - Distributed AI Teams (Experimental)​

Connect and orchestrate CrewX instances across projects and servers:

  • Cross-project experts - Frontend dev asks backend team's API specialist agent
  • Team collaboration - Each team builds their own agents, entire org can use them
  • Expert knowledge sharing - Ask senior's code review agent, security team's audit agent anytime
  • Separate but connected - Each project keeps its own context, collaborate when needed
# Access another project's specialized agents
providers:
  - id: backend_project
    type: remote
    location: "file:///workspace/backend-api/crewx.yaml"
    external_agent_id: "api_expert"

# Use their expertise in your project
crewx query "@api_expert design user authentication API"
crewx execute "@api_expert implement OAuth flow"

Plugin Provider System - Universal AI Integration​

Transform any CLI tool or AI service into an agent:

  • Bring Your Own AI - OpenAI, Anthropic, Ollama, LiteLLM, or any AI service
  • Bring Your Own Tools - jq, curl, ffmpeg, or any CLI tool becomes an agent
  • Bring Your Own Framework - Integrate LangChain, CrewAI, AutoGPT seamlessly
  • No coding required - Simple YAML configuration
  • Mix and match - Combine different AI services in one workflow
# Example: Add any AI service as a plugin
providers:
  - id: ollama
    type: plugin
    cli_command: ollama
    default_model: "llama3"
    query_args: ["run", "{model}"]
    prompt_in_args: false

agents:
  - id: "local_llama"
    provider: "plugin/ollama"

Project Scaffolding - Template System​

Download and install project templates with pre-configured agents and documents:

  • Instant setup - Install ready-to-use agent configurations
  • Team sharing - Package and distribute agent setups across projects
  • Template management - List, install, and manage templates via CLI
  • Quick start - Get started with proven agent configurations
# List available templates
crewx template list

# Install WBS automation template
crewx template init wbs-automation

# Install Docusaurus i18n template
crewx template init docusaurus-i18n

# Check installed agents
crewx agent ls

👉 Project Scaffolding Guide → for detailed usage

Claude Skills Compatible - Reusable AI Expertise​

Share and reuse specialized AI capabilities using Claude Code skills format:

  • 100% Claude Code compatible - Use existing Claude skills without modification
  • Agent enhancement - Add specialized capabilities to any agent
  • Progressive disclosure - Skills load metadata first, content on-demand
  • Cross-agent sharing - One skill, multiple agents
  • Simple YAML + Markdown - Easy to create and maintain
# Enable skills for your agents
skills:
  paths:
    - ./skills                    # Custom skill directories
  include:
    - hello                       # Specific skills to load
    - code-reviewer
    - api-designer

agents:
  - id: "senior_dev"
    provider: "cli/claude"
    skills:
      include:
        - code-reviewer           # Agent-specific skills
        - api-designer
    inline:
      prompt: |
        You are a senior developer with specialized skills.

👉 Skills System Guide → for detailed usage

Other Benefits​

  • No additional costs - Use existing Claude Pro, Gemini, Codex or GitHub Copilot subscriptions
  • Multi-agent collaboration - Different AI models working on specialized tasks
  • Parallel execution - Multiple agents working simultaneously
  • Flexible integration - CLI, MCP server, or Slack bot

Quick Start​

🚀 One-Command Setup (Recommended)​

# Interactive setup - creates crewx.yaml and helps with Slack setup
npx crewx-quickstart

📦 Manual Installation​

# Install
npm install -g crewx

# Initialize
crewx init

# Check system
crewx doctor

# Try it out
crewx query "@claude analyze my code"
crewx execute "@claude create a login component"

Three Ways to Use​

Slack Mode - Team Collaboration (Recommended)​

# Start CrewX in your Slack workspace (read-only query mode)
crewx slack

# Allow agents to run execute tasks (file changes, migrations, etc.)
crewx slack --mode execute

# Your team can now:
# - @mention AI agents in channels
# - Maintain context in threads
# - Share AI insights with the whole team

👉 Complete Slack Setup Guide →

CLI Mode - Direct terminal usage​

crewx query "@claude review this code"
crewx execute "@gemini optimize performance"
crewx query "@claude @gemini @copilot compare approaches"

MCP Server Mode - IDE integration​

crewx mcp  # VS Code, Claude Desktop, Cursor

Supported AI Tools​

CLI Providers (Local)​

  • Claude Code - Advanced reasoning and analysis
  • Gemini CLI - Real-time web access
  • GitHub Copilot CLI - Specialized coding assistant
  • Codex CLI - Open inference with workspace-aware execution

API Providers (Cloud & Local)​

CrewX supports unified HTTP-based API providers including:

  • Ollama, OpenRouter, Anthropic, OpenAI
  • Support for LiteLLM, Google AI, AWS Bedrock, and more

👉 Complete API Provider Guide → for detailed setup and examples

Basic Usage​

# Read-only analysis
crewx query "@claude explain this function"

# File creation/modification
crewx execute "@claude implement user authentication"

# Parallel tasks
crewx execute "@claude create tests" "@gemini write docs"

# Pipeline workflows
crewx query "@architect design API" | \
crewx execute "@backend implement it"

# Thread-based conversations
crewx query "@claude design login" --thread "auth-feature"
crewx execute "@gemini implement it" --thread "auth-feature"

# Codex CLI agent
crewx query "@codex draft a release checklist"

Built-in CLI providers:

  • cli/claude
  • cli/gemini
  • cli/copilot
  • cli/codex

Create Custom Agents​

# Let SowonAI CrewX create agents for you
crewx execute "@crewx Create a Python expert agent"
crewx execute "@crewx Create a React specialist with TypeScript"
crewx execute "@crewx Create a DevOps agent for Docker"

# Test your new agent
crewx query "@python_expert Review my code"

Agent Configuration​

Create crewx.yaml (or agents.yaml for backward compatibility):

agents:
  - id: "frontend_dev"
    name: "React Expert"
    working_directory: "./src"
    inline:
      type: "agent"
      provider: "cli/claude"  # Built-in CLI provider
      prompt: |
        You are a senior React developer.
        Provide detailed examples and best practices.

Note: crewx.yaml is the preferred configuration file name. The legacy agents.yaml is still supported for backward compatibility. If both files exist, crewx.yaml takes priority.

Layout System​

CrewX layouts provide reusable prompt templates that separate structure from content.

Quick Example​

# crewx.yaml
agents:
  - id: full_agent
    inline:
      layout: "crewx/default"  # Full agent profile
      prompt: |
        You are a comprehensive assistant.

  - id: simple_agent
    inline:
      layout: "crewx/minimal"  # Lightweight wrapper
      prompt: |
        You are a simple assistant.

Features:

  • 🎨 Reusable Templates - Share layouts across agents
  • ⚛️ Props Schema - React PropTypes-style validation for custom layouts
  • 🔧 Built-in Layouts - crewx/default, crewx/minimal
  • 🛡️ Security Containers - Automatic prompt wrapping

👉 Layout System Guide → for detailed usage

Remote Agents (Experimental)​

Connect to other CrewX instances and delegate tasks across projects or servers.

Quick Example:

# Add a remote CrewX instance
providers:
  - id: backend_server
    type: remote
    location: "http://api.example.com:3000"
    external_agent_id: "backend_team"

agents:
  - id: "remote_backend"
    provider: "remote/backend_server"

# Use it like any other agent
crewx query "@remote_backend check API status"

Use Cases:

  • Project isolation - Separate configurations for different codebases
  • Distributed teams - Each team runs their own CrewX with specialized agents
  • Resource sharing - Access powerful compute resources remotely
  • Multi-project coordination - Orchestrate work across multiple projects

👉 Remote Agents Guide → for detailed setup and configuration

Documentation​

User Guides​

Developer Guides​

License​

  • SDK (@sowonai/crewx-sdk): Apache-2.0 License
  • CLI (crewx): MIT License

Copyright (c) 2025 SowonLabs

Contributing​

We welcome contributions! Please read our Contributing Guide before submitting pull requests.

Built by SowonLabs