🤖 AI Coding Tools & Marketplace Integration

Connect your favorite AI coding assistant directly to Chatty's 55-tool Model Context Protocol (MCP) server.

Manage chatbots, generate visual flows, audit performance, and review lead analytics directly from your conversational coding workflow without switching context.

⚡ 1. Claude Code CLI Integration

Chatty publishes an official Claude Code plugin manifest under .claude-plugin/marketplace.json.

Add Marketplace:

Inside your Claude Code session, register Chatty:

/plugin marketplace add PersonaliAI/chatty

Or install directly:

claude plugin install chatty

Once installed, ask Claude:

"Audit my Chatty bot 8f9024b1-e25c-4122-8d77-a82a6fce921b and show the 5-pillar scorecard."
"Create a new customer support chatbot for my store and seed it with templates/ecommerce-store.json."

💻 2. Cursor IDE Integration

Cursor supports both native MCP servers and repository rules.

Step 1: Configure MCP in Cursor

1. Open Cursor Settings (Cmd + , or Ctrl + ,).
2. Go to Features → MCP Servers → Add New MCP Server.
3. Fill in:
   • Name: chatty
   • Type: sse
   • Server URL: https://your-backend-domain/mcp (or http://localhost:8000/mcp)
4. Authorize via the OAuth consent popup on first connect.

Step 2: Cursor Rules Active

The Chatty repository includes:

• .cursor/rules/chatty.mdc: Automatic system guidance on Chatty architecture, tools, and security.
• .cursorrules: Universal editor prompt rules.

Ask Cursor's composer (Ctrl + I or Cmd + I):

"Use @chatty to generate a 4-step lead qualification flow and run dry-run validation."

🖥️ 3. Claude Desktop Integration

Edit your Claude Desktop configuration file:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

Add the Chatty MCP server:

{
  "mcpServers": {
    "chatty": {
      "url": "https://api.chatty.personaliai.com/mcp"
    }
  }
}

Restart Claude Desktop. A browser tab will open asking you to authorize your Chatty account. Once approved, all 55 tools are live.

🏄 4. Windsurf / Cascade Integration

In ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "chatty": {
      "serverUrl": "https://api.chatty.personaliai.com/mcp"
    }
  }
}

Cascade can now inspect live chatbot configurations, test RAG semantic search, and audit conversion rates during editor sessions.