Twenty CRM

Twenty CRM is an MCP server that provides an interface between Model Context Protocol clients and the Twenty CRM platform. It connects directly to the Twenty CRM API, allowing sales teams, customer success representatives, and operations managers to interact with CRM records through conversational AI tools. By exposing Twenty CRM's data endpoints over the Model Context Protocol, the server allows language models to read, create, and update customer profiles, business accounts, pipeline stages, and communication histories. Users who rely on Twenty for customer relationship management can streamline daily administration without manually switching browser tabs or navigating complex database forms. The integration enables AI agents to retrieve background details on prospects prior to calls, log interaction notes directly from conversation threads, and review active deal stages in real time. Organizations hosting or using Twenty CRM can maintain centralized records while giving client applications standard tool-based access to perform routine queries and updates across their customer records.

Category: CRM, ERP & E-commerce

Tags: crm, customer-data, sales, twenty

Visit Twenty CRM

How to install and configure Twenty CRM

1. Review the installation prerequisites and clone the project from the official repository at https://github.com/OleApp-de/TwentyMCP. 2. Consult the repository documentation for required environment variables, including your Twenty CRM instance URL and API access token. 3. Build or set up the project locally as specified in the repository guidelines. 4. Open your MCP client configuration file (such as the Claude Desktop or Cursor configuration JSON). 5. Add an entry under the mcpServers block pointing to the server binary or startup script with the required credentials. 6. Restart your MCP client to initialize the server connection.

What you can do with Twenty CRM

• Querying contact and company records directly from chat to check account status, recent correspondence, and assigned team members before client meetings. - Creating new sales leads and customer accounts in Twenty CRM using structured data extracted from conversational notes or meeting summaries. - Updating deal stages, opportunity values, and close dates within the sales pipeline based on prompt instructions from account executives. - Searching customer interaction logs and activity timelines to answer historical context questions during ongoing client support operations. - Automating task creation and follow-up reminders inside Twenty CRM directly through conversational commands issued to an AI assistant.

Key facts

• https://github.com/OleApp-de/TwentyMCP
• CRM, ERP & E-commerce, Project & Task Management
• crm, customer-data, sales, twenty

Part of MCP Servers

What is Twenty CRM MCP?

Twenty CRM MCP is a Model Context Protocol server that links AI client applications with the Twenty CRM platform. It lets conversational agents access the Twenty CRM API to retrieve customer information, search through contact databases, and update sales pipeline entries using standard tool calls.

What can Twenty CRM do?

The server enables AI assistants to interact with customer relationship data stored in Twenty CRM. Users can search for accounts, inspect contact records, log interaction notes, create new prospect entries, and update existing deals or pipeline stages directly through AI chat prompts without opening the web interface.

Which MCP clients work with Twenty CRM?

Any client application that supports the Model Context Protocol can communicate with Twenty CRM. Compatible environments include Claude Desktop, Cursor, and custom agent frameworks configured with MCP client libraries, provided the client has network access and appropriate API credentials to communicate with the server.

How do I install Twenty CRM?

Installation instructions are maintained in the project repository on GitHub. Because setup details depend on the implementation, you should inspect the repository documentation at https://github.com/OleApp-de/TwentyMCP for prerequisites, required Twenty API keys, runtime requirements, and configuration snippets needed to register it with your client.