🤖 Twenty CRM MCP Server
Transform your CRM into an AI-powered assistant
Power your agent
Instal dari penerbit. Bukan endpoint PlugMCP.
URL remote atau perintah di mesin Anda? Perbedaan kedua cara instal
Jalankan di PlugMCP
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
BukaDari penerbit
A Model Context Protocol (MCP) server for Twenty CRM integration. Enables natural language interactions with your CRM data through Claude and other AI assistants. Supports CRUD operations, dynamic schema discovery, and advanced search across people, companies, tasks, and notes.
A Model Context Protocol server that connects Twenty CRM with Claude and other AI assistants, enabling natural language interactions with your customer data.
filter grammar (ilike matching), not the nonexistent search paramfirstName/email/amount are converted to Twenty's name/emails/amount composite shapes on writestartingAfter/endingBefore) plus raw filter, orderBy and depth passthrough on every list toolpersonId/companyId/opportunityId when creating and the target links are created for youRetry-After), structured error messagesGet your Twenty CRM API key: in Twenty, go to Settings → API & Webhooks (under Developers) and generate a key.
Configure your MCP client.
Claude Desktop (claude_desktop_config.json):
{
"mcpServers": {
"twenty-crm": {
"command": "npx",
"args": ["-y", "github:mhenry3164/twenty-crm-mcp-server"],
"env": {
"TWENTY_API_KEY": "your_api_key_here",
"TWENTY_BASE_URL": "https://api.twenty.com"
}
}
}
}
Claude Code:
claude mcp add twenty-crm -e TWENTY_API_KEY=your_api_key_here -e TWENTY_BASE_URL=https://api.twenty.com -- npx -y github:mhenry3164/twenty-crm-mcp-server
Or clone and run from source:
git clone https://github.com/mhenry3164/twenty-crm-mcp-server.git
cd twenty-crm-mcp-server
npm install
then point your client at node /path/to/twenty-crm-mcp-server/index.js.
Once configured, use natural language to interact with your Twenty CRM:
"List the first 10 people in my CRM"
"Create a new person named John Doe with email [email protected] at Acme (company id ...)"
"Create a $12,000 opportunity called 'Acme renewal' in PROPOSAL stage"
"Add a note about today's call and attach it to the Acme company record"
"Create a follow-up task due Friday assigned to Sam, linked to the deal"
"Search people, companies and opportunities for 'blockchain'"
Twenty stores identity/contact/money data in composite fields. This server accepts flat parameters and maps them to the shapes the API expects:
Anything else — including your workspace's custom fields — passes through untouched.
Note on company revenue: the field is
annualRecurringRevenueon most deployed Twenty versions andannualRevenueon Twenty v2.x+. Both parameters are supported; use the one that matches your instance (checkget_object_metadataforcompaniesif unsure).
Twenty's REST API has no search or offset query params — this server builds real filters instead:
search on list tools becomes a case-insensitive ilike filter across sensible fields per object (people: first/last name + email; companies: name + domain; opportunities/notes/tasks: name/title).filter accepts Twenty's raw filter grammar for anything more specific:
createdAt[gte]:2026-01-01or(city[eq]:Austin,city[eq]:Dallas)emails.primaryEmail[ilike]:%@acme.comeq, neq, in, containsAny, is, gt, gte, lt, lte, startsWith, endsWith, like, ilike; combine with and(...), or(...), not(...). Use field[is]:NULL / NOT_NULL for empty checks.orderBy: e.g. createdAt[DescNullsLast] (directions: AscNullsFirst, AscNullsLast, DescNullsFirst, DescNullsLast).pageInfo.endCursor — pass it as startingAfter for the next page. limit max is 200.depth: 0 (default) or 1 to include first-level related records.docker build -t twenty-crm-mcp-server .
{
"mcpServers": {
"twenty-crm": {
"command": "docker",
"args": ["run", "-i", "--rm", "-e", "TWENTY_API_KEY", "-e", "TWENTY_BASE_URL", "twenty-crm-mcp-server"],
"env": {
"TWENTY_API_KEY": "your_api_key_here",
"TWENTY_BASE_URL": "https://api.twenty.com"
}
}
}
}
npm install
npm test # unit tests (node --test, no network needed)
npm start # run the stdio server
v2.0.0 fixes writes and search against the actual Twenty API. Breaking changes:
offset was removed from list tools (Twenty's REST API never supported it) — use cursor pagination (startingAfter).body now maps to bodyV2.markdown; position params were dropped.PATCH (Twenty's documented verb) instead of PUT.{ records, pageInfo, totalCount? } instead of the raw API envelope.This project is community-supported — the maintainer no longer runs Twenty day-to-day, so bug reports and PRs from active Twenty users are especially welcome and get reviewed and merged. See CONTRIBUTING.md.
v2.0.0 builds on excellent community work: composite-field mapping (#4 by @zaks), filter-based search (#6 by @studio7A, #8 by @InDebted-Growth), opportunity tools (#7 by @tarikhennen), and patterns from the forks by @BCJonkhout, @archie1492, @ndrkltsk and @atilladeniz.
MIT — see LICENSE.
A Model Context Protocol (MCP) server for Twenty CRM integration. Enables natural language interactions with your CRM data through Claude and other AI assistants. Supports CRUD operations, dynamic sch
Ringkasan umum
A Model Context Protocol (MCP) server for Twenty CRM integration. Enables natural language interactions with your CRM data through Claude and other AI assistants. Supports CRUD operations, dynamic schema discovery, and advanced search across people, companies, tasks, and notes.
Tool
Belum ada daftar tool. Server remote mendapatkannya setelah pemeriksaan kesehatan pertama.
FAQ
Salin konfigurasi stdio untuk Cursor, Claude Desktop, Claude Code, atau VS Code di halaman ini. Konfigurasi ini memakai instalasi dari penerbit, bukan proxy PlugMCP.
Tidak ada variabel lingkungan wajib yang tercantum di catatan ini.
Twenty Crm berjalan secara lokal melalui stdio.
Terakhir diperiksa 2026-11-05. A Model Context Protocol (MCP) server for Twenty CRM integration. Enables natural language interactions with your CRM data through Claude and other AI assistants. Supports CRUD operations, dynamic schema discovery, and advanced search across people, companies, tasks, and notes.