MCP Suite CRM
Enables AI agents to interact with a CRM covering companies, people, leads, deals, and more, with role checks, scoped agent keys, approval gates, and a shared audit trail.
README
MCP Suite CRM
An open-source CRM built for humans and AI agents to operate together. The web application and every MCP client use the same typed operation catalog, with role checks, scoped agent keys, approval gates for risky agent actions, and a shared audit trail.
Pre-release: the source is public, but the supported installer, npm package, container image, and stable release are not published yet.
What is included
| Path | Purpose |
|---|---|
apps/web |
Login, the authenticated CRM under /app/*, and the HTTP operation API |
apps/mcp |
MCP over stdio and HTTP |
packages/core |
Domain types, authorization policy, and operation catalog |
packages/db |
SQLite persistence, schema, authentication, and bootstrap |
The CRM covers companies, people, leads, deals, activities, tasks, pipelines, tags, contact lists, custom fields, saved views, offerings, approvals, users, agents, and audit events. The public marketing site, signup, billing, and managed-hosting controls are deliberately not part of this repository.
Run from source
The supported development platform is 64-bit Linux. Install mise; the repository pins Node 22 and pnpm 10.
mise trust
mise exec -- make setup
mise exec -- make dev
Open http://localhost:2222. / redirects to /login. The first setup creates
the SQLite database at data/emcp.db and prints the initial owner credentials;
change the password after signing in.
Run with Docker
docker compose up -d
One container, one process on :2222: the web app, the operation API, the
MCP endpoint for agents (POST /mcp) and GET /healthz. Persistent state
lives in ./data. The compose file builds the mcpsuite/crm image locally
until official images are released. See PRODUCTION.md for
the standalone-MCP command override and reverse-proxy guidance.
Connect an agent
In the CRM, open Admin → Agents, create an agent client, and copy the key shown once. The UI provides a ready-to-paste configuration. The full transport, scope, and trust-profile guide is in docs/mcp.md.
Example for Claude Code:
claude mcp add --transport http mcpsuite-crm http://localhost:2222/mcp \
--header "Authorization: Bearer emcp_YOUR_KEY"
Development checks
mise exec -- make typecheck
mise exec -- make test
mise exec -- make build
mise exec -- make smoke
See AGENTS.md for the code architecture and conventions, PRODUCTION.md for the current self-hosting runbook, and CONTRIBUTING.md before opening a pull request.
License
AGPL-3.0-only. See LICENSE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.