OpenAgent - Open-source AI development assistant

ℹ️ IMPORTANT: Name Change Notice

This project was previously named "OpenClaude" but has been rebranded to "OpenAgent" to avoid trademark conflicts with Anthropic's Claude brand. All functionality remains the same - we simply changed the name to ensure complete legal compliance and professional distribution.

⚠️ UNDER ACTIVE DEVELOPMENT - IN BETA

SOME FEATURES MAY NOT WORK AS EXPECTED

---

✨ Features

🧠 Smart Context Management - Remembers your project state across sessions
⚡ Token Optimization - Save costs with intelligent compression
🛡️ Code Validation - Multi-layer quality checks for your code
🔧 MCP Integration - Use any MCP server for extended functionality
💾 Persistent Memory - Learn from your interactions and preferences
🚀 Real-time Streaming - Professional interface with tool visibility

---

📖 Project Documentation

📋 Complete Code Summary

For developers who want to understand the entire OpenAgent codebase quickly, we provide a comprehensive code summary:

📄 CODE_SUMMERY.md - Complete project architecture, file relationships, and development guide

This document includes:

• 🏗️ Complete Architecture Overview - All 8 system layers explained
• 🔗 File Relationships - How every file connects and interacts
• 📊 Code Quality Analysis - Health metrics, dead code analysis, optimization opportunities
• 🎯 Developer Onboarding - Quick start checklist for new contributors
• 🛠️ Modification Guide - Exactly which files to change for different types of updates
• 📈 Performance Metrics - Project health score, token optimization insights

---

🚀 Quick Start

# Clone and setup
git clone https://github.com/RaheesAhmed/OpenAgent.git
cd OpenAgent
npm install
npm run build

# Set your API key
export ANTHROPIC_API_KEY="your-key-here"

# Start using OpenAgent
node ./dist/index.js

---

🎮 Slash Commands

OpenAgent includes a powerful set of slash commands for controlling your development session:

📋 Core Commands

• /help or /h or /? - Show all available commands
• /status or /st - Display agent status, metrics, and performance data
• /version or /v - Show OpenAgent version and model information
• /exit or /quit or /q - Clean shutdown and exit

🔄 Session Management

• /reset or /clear or /restart - Reset agent and clear conversation history
• /new or /thread - Create a new conversation thread
• /history [threadId] or /hist - Show conversation history for current or specified thread
• /metrics or /stats - Display detailed performance metrics

🛠️ Development Tools

• /revert [steps] or /rv - Actually revert file changes (deletes created files, undoes edits)
• /undo or /u - Undo the last agent action (same as /revert 1)
• /diff [file] or /d - Show all changes made by the agent with file previews
• /config or /cfg - View/update agent configuration

💾 State Management

• /checkpoint create <name> or /cp create - Create a restore point
• /checkpoint list or /cp list - List all available checkpoints
• /checkpoint restore <name> or /cp restore - Restore from saved checkpoint
• /rollback [checkpoint] or /rb - Rollback to previous stable state

💡 Usage Examples

# Check what the agent has been doing
/status

# See all file changes made
/diff

# Undo the last change (actually deletes/reverts files)
/undo

# Revert last 3 operations
/revert 3

# Create a checkpoint before major changes
/checkpoint create before-refactor

# View conversation history
/history

# Start fresh
/reset

🔥 Real File Operations: Unlike other AI assistants, OpenAgent's /revert and /undo commands actually modify your filesystem - deleting created files and reversing edits by analyzing the conversation history.

---

⚙️ Configuration

OpenAgent creates .openagent/ folder with configuration files:

Model Configuration

Edit .openagent/config.json to use any supported model:

{
  "model": "anthropic:claude-sonnet-4-20250514",
  "temperature": 0.1,
  "maxTokens": 4000,
  "memoryEnabled": true,
  "streamingEnabled": true,
  "humanInLoopEnabled": true,
  "checkpointsEnabled": true
}

Supported models:

• anthropic:claude-sonnet-4-20250514 (recommended)
• anthropic:claude-3-5-sonnet-20241022
• anthropic:claude-3-opus-20240229
• openai:gpt-4
• openai:gpt-3.5-turbo
• Any other LangChain universal model

MCP Server Configuration

Add MCP servers in .openagent/mcp-servers.json:

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/allowed/files"]
    },
    "context-engine": {
      "command": "npx", 
      "args": ["-y", "context-engine-mcp"]
    },
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "${GITHUB_TOKEN}"
      }
    }
  }
}

Custom AI Rules

Set project-specific rules in .openagent/rules.md:

# Project Rules

## Code Standards
- Use TypeScript for all new files
- Follow existing naming conventions
- Add JSDoc comments for public APIs

## Architecture
- Keep components small and focused
- Use proper error handling
- Write tests for new features

## Security
- Never commit secrets or API keys
- Validate all user inputs
- Use secure coding practices

---

💬 Usage Examples

# Basic conversation
You: Create a React login component

# File operations  
You: Read the package.json file
You: Create a new utils.ts file with helper functions

# Development tasks
You: Help debug this TypeScript error
You: Run npm test and show results
You: Optimize this SQL query for better performance

What you'll see:

⠋ OpenAgent is thinking (2s)...
🔧 read_file
🔧 create_file

✅ Created React login component with validation
💰 Cost: $0.0045 (1200 + 300 tokens)
📊 Tokens: 1200 in + 300 out = 1500 total

---

🏗️ Project Structure

.openagent/
├── config.json           # Model and behavior configuration
├── mcp-servers.json      # MCP server setup
├── rules.md              # Custom AI rules and guidelines
├── memory/               # Session and learning data
└── logs/                 # Operation logs

src/
├── agents/               # AI agent implementation
├── cli/                  # Command-line interface
├── core/                 # Context, memory, validation
├── mcp/                  # MCP client integration
└── tools/                # Built-in tools

---

🔧 Built-in Capabilities

File Operations (via MCP)

• Read files - View content with syntax highlighting
• Create/Edit files - Generate or modify code files
• Search & Replace - Pattern-based text editing
• Directory listing - Browse project structure

Development Tools

• Terminal integration - Execute commands safely
• Code validation - Check syntax and quality
• Memory system - Learn your preferences
• Cost tracking - Monitor API usage

---

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for detailed instructions.

Quick steps:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if needed
5. Submit a pull request

Development Setup

git clone https://github.com/RaheesAhmed/OpenAgent.git
cd OpenAgent
npm install
npm run dev

---

📄 License

MIT License - see LICENSE for details.

---

🐛 Report Bug • 💡 Request Feature

OpenAgent - Your configurable AI development assistant