Quick answer: Model Context Protocol (MCP) is a new Leadtrekker feature that connects your live lead data directly to AI assistants like Claude and ChatGPT. Instead of manually exporting reports, you can now ask your AI questions in plain English, such as “How many leads came from Google Ads this month?” or “Show me pending leads assigned to Y”, and get instant, real-time answers based strictly on your account permissions.
Let’s explore this new feature in more depth below.
What is MCP?
MCP (Model Context Protocol) is an open standard that lets AI assistants connect directly to the tools you already use. Instead of copying data out of Leadtrekker and pasting it into ChatGPT, the assistant queries Leadtrekker itself and answers from live data.
Think of it as giving your AI assistant a read-only key to your Lead Management System. One that respects exactly the same permissions you have when you log in.
What you can do with MCP
Once connected, you ask questions in plain language:
- “How many leads did we get this month?”
- “Show me everything assigned to Y that’s still pending.”
- “Which leads came from Google Ads in September?”
- “What’s in my pipeline from Facebook that we’ve marked as lost?”
Your assistant can then summarise, compare, or draft follow-ups from the results, without you building a single report.
Security and permissions
- You sign in with your own Leadtrekker credentials. Your password is never shared with OpenAI or Anthropic.
- The connection respects your Leadtrekker permissions. If you can only see your own leads in Leadtrekker, that’s all your assistant sees.
- Access is read-only. Nothing in your account can be changed.
- You can disconnect at any time.
Before you start
Sign in to Leadtrekker at least once in a browser. If you haven’t signed in recently, the connection will fail until you do.
Your MCP server URL: https://mcp.leadtrekker.com/mcp/leadtrekker
Connecting with Claude
Custom connectors are available on all Claude plans — Free, Pro, Max, Team and Enterprise. Free accounts are limited to one custom connector.
- Open Claude in a web browser and go to Settings → Connectors
- Click Add custom connector
- Name it Leadtrekker and paste the server URL above
- Leave the Advanced settings (OAuth Client ID and Secret) empty
- Click Add — Claude opens a Leadtrekker sign-in page
- Sign in with your Leadtrekker email and password, then approve access
- In a new chat, click the + button, hover Connectors, and toggle Leadtrekker on
On Team and Enterprise plans, an Owner must add the connector for the organisation first under Organisation settings → Connectors. Members then connect individually.
Connecting with ChatGPT
ChatGPT needs one extra step: Developer Mode.
A paid plan is required — Plus, Pro, Business, Enterprise or Edu. Developer Mode is only available at chatgpt.com in a browser, not in the mobile app.
Enable Developer Mode (once per account)
- Open ChatGPT in a web browser
- Go to Settings, then Connectors (some accounts label this Apps & Connectors)
- Open Advanced settings and turn on Developer mode
Add the connector
- Still under Settings → Connectors, click Create (or + / New connector)
- Name it Leadtrekker and paste the server URL above
- Choose OAuth for authentication, then save
- Sign in with your Leadtrekker credentials and approve access
- In a new chat, enable Leadtrekker from the composer’s tools menu
If the connector shows “Disconnected” afterwards, that’s normal. ChatGPT only connects when it actually needs the tools during a chat.
On Business, Enterprise and Edu workspaces this is admin-controlled: an Owner or Admin must enable Developer Mode and custom apps in Workspace settings.
Troubleshooting
- Sign-in fails — sign in to Leadtrekker directly first, then try again
- No “Add custom connector” option in ChatGPT — check your plan and that Developer Mode is switched on
- Tools missing after an update — disconnect and reconnect the connector
Conclusion
Model Context Protocol (MCP) bridges the gap between your lead management system and conversational AI. By giving Claude or ChatGPT secure, read-only access to Leadtrekker, you eliminate the need to manually build custom reports or export spreadsheets.
Whether you need a quick headcount on this month’s Google Ads leads or want to analyse lost pipeline opportunities on Facebook, MCP lets you query your live data in plain English while respecting your team’s existing account permissions.
Experience Smarter Lead Management with Leadtrekker
Want to ask your AI assistant direct questions about your pipeline without exporting a single file?
Try Leadtrekker’s new MCP integration and turn your sales data into actionable business intelligence instantly.