Connect ByteRover to your favorite AI coding assistant and give it persistent memory across all your sessions.

Overview

ByteRover works with your existing AI tools through the Model Context Protocol (MCP), enabling them to remember code patterns, solutions, and context across conversations. Setup takes just a few minutes.

Supported AI Assistants

Standalone Tools

IDE Extensions

These extensions work within Cursor, VS Code, or Windsurf. Select the extension, then choose your IDE for automatic setup.

Quick Setup Guide

ByteRover Quick Setup Process: Choose Your Tool, Connect Your Assistant, Verify Connection

Step 1: Choose Your Tool

ByteRover supports two connection types: Automatic Setup (Recommended)
  • Cursor, Windsurf, Trae.ai, VS Code Extensions
  • One-click connection through ByteRover dashboard
  • API keys generated automatically
Manual Configuration
  • Claude Code, Claude.ai, Qwen Code, Gemini CLI, Goose, Zed
  • Requires adding MCP server configuration
  • Takes 2-3 minutes to set up

Step 2: Connect Your Assistant

  1. Open your ByteRover dashboard
  2. Click “Connect your agent”
  3. Select your tool (e.g., Cursor)
  4. Click “Connect”
  5. Approve the extension installation when prompted
That’s it! Your AI assistant now has memory.

Step 3: Test Your Connection

Try this simple test to verify everything works:
Store this utility function to ByteRover:

function debounce(func, wait) {
  let timeout;
  return function executedFunction(...args) {
    const later = () => {
      clearTimeout(timeout);
      func(...args);
    };
    clearTimeout(timeout);
    timeout = setTimeout(later, wait);
  };
}
If connected properly, your AI will confirm the function was saved.

How It Works

Creating Memories

Ask your AI to store useful code patterns:
Store this React hook for API calls to ByteRover:

function useApi(endpoint) {
  const [data, setData] = useState(null);
  const [loading, setLoading] = useState(true);
  const [error, setError] = useState(null);
  
  useEffect(() => {
    fetch(endpoint)
      .then(res => res.json())
      .then(setData)
      .catch(setError)
      .finally(() => setLoading(false));
  }, [endpoint]);
  
  return { data, loading, error };
}
Your AI automatically saves it to your ByteRover knowledge base.

Retrieving Memories

Later, in any conversation, ask:
Search my memory for that React API hook pattern
Your AI recalls the exact pattern with full context, even across different sessions or projects.

Usage & Pricing

PlanMonthly RetrievalsPrice
Hobby200Free
Pro1,000$10/mo
Team15,000Custom
Each memory search counts as one retrieval. Usage resets monthly.

Best Practices

Storing Memories

✅ "Store this Express.js authentication middleware pattern"
❌ "Remember this"

Searching Memories

✅ "Find my React error boundary implementations"
❌ "Get that thing from before"

Common Use Cases

  • Code Patterns: Store reusable components, hooks, utilities
  • Architecture Decisions: Document why certain choices were made
  • Bug Solutions: Remember fixes for tricky issues
  • API Integrations: Save working connection patterns
  • Configuration: Keep track of complex setup procedures

Need Help?

  • Documentation: Full guides for each tool in /quickstart/[tool-name]
  • Discord Community: Join for support
  • Dashboard: Manage connections and view usage at byterover.dev