Connie

Connie MCP server

Connie is an AI workspace for go-to-market teams: a CRM, Lists, a multichannel inbox and workflow automation in one place. The Connie MCP server connects your AI assistant (GitHub Copilot in VS Code, Claude, ChatGPT, Cursor and any other MCP client) to your Connie workspace, so it can work with your Records, Lists and workflows using your own permissions.

This repository describes the server. The server itself is hosted by Connie; there is nothing to install or run locally.

• Server URL: https://mcp.connie.ai
• Transport: Streamable HTTP
• Authentication: OAuth 2.0 with automatic client registration
• Registry name: ai.connie/connie
• Full setup guide: https://connie.ai/docs/mcp

What it can do

The server exposes five tools:

Through use-tool, your assistant can reach:

• Records and Lists: search, read, create, update and delete People, Companies, Deals, Tasks and other Records, and manage Lists.
• CSV import and export into and out of your Lists.
• Lead search and enrichment: find new people and companies, and add details such as email addresses, phone numbers and company information to your Records.
• Workflows: build, edit and run them.
• Sending: email and messages from the accounts connected to your workspace.
• Connected apps: actions in the apps your workspace has connected, such as your CRM, Slack or outreach tools. These read from and write to those apps, outside your Connie workspace.
• Your workspace: workspace details, members, files and skills.

Scope, permissions and credits

• The server acts only in the Connie workspaces you are a member of, and every tool runs with your own permissions. A guest in a workspace cannot write through it.
• Actions in connected apps use the accounts connected to that workspace. A tool turned off in the workspace's AI tool settings cannot run through the server.
• Tools that enrich or research consume credits from the workspace's balance, as they do in the Connie app. Every result includes a billing object with the credits actually charged.

Setup

You need a Connie account (connie.ai) and membership of at least one workspace. The first time your client connects, your browser opens a Connie sign-in and consent screen; approve it and the client is connected.

VS Code (GitHub Copilot)

Add the server to .vscode/mcp.json in your project, or to your user configuration with MCP: Open User Configuration:

{
  "servers": {
    "connie": {
      "type": "http",
      "url": "https://mcp.connie.ai"
    }
  }
}

Start the server from the editor, complete the Connie sign-in when VS Code asks you to authenticate, then use Copilot Chat in agent mode.

Claude Code

claude mcp add --transport http connie https://mcp.connie.ai

Then run /mcp in Claude Code and select Connie to sign in.

Claude, ChatGPT and other clients

Add a custom connector or remote MCP server with the URL https://mcp.connie.ai and sign in when asked. Step-by-step instructions for Claude, ChatGPT, Codex, Cursor, Grok and other clients are in the setup guide.

Using it

Ask in plain language, for example:

• "List my Connie workspaces."
• "Find every company in the Warm Leads list without a website."
• "Import this CSV into a new List called Event Attendees."
• "Find the work email for the people in my Q4 Prospects list."
• "Create a workflow that adds new HubSpot contacts to my Inbound list."

Your assistant finds the right tool with search-tools, checks what it needs with get-tool, and runs it with use-tool.

Managing access

In Connie, open Connectors in the left sidebar and the MCP view. Each client you have authorized appears as a card; revoke one, or use Revoke all to disconnect a client everywhere. Revocation is checked on every request.

Privacy and support

• Lead search and enrichment can return personal contact details sourced from third-party business data providers and public sources. They are stored only in your workspace. See the Privacy Policy for where this information comes from and how someone can ask for it to be removed.
• Support: [email protected]