OpenTeleCRM MCP Server

OpenTeleCRM MCP Server

Enables interaction with a telecalling-first sales CRM, providing 13 JSON-RPC MCP tools for lead management, actions, WhatsApp, and telephony, with multi-tenant RLS.

Category
Visit Server

README

OpenTeleCRM

1:1 FOSS clone of TeleCRM — a telecalling-first sales CRM. Multi-tenant from line one, self-hosted natively (no Docker), PostgreSQL with RLS.

Surface Status Details
REST API ✅ 62/62 tests NestJS + Fastify on port 3005
JSON-RPC MCP ✅ 13 tools Streamable HTTP on port 3006
Bruno collection ✅ 27 reqs, 58 assertions TeleCRM wire-compatible
WhatsApp ✅ wwebjs driver Pairing-code auth, templates, broadcasts
Telephony ✅ Dialer + callbacks OSS provider on Asterisk

Quick start

make setup      # provision deps + install + db-init + db-migrate + db-seed
pnpm dev        # API on :3005, MCP on :3006
pnpm test       # 62 contract tests, 7 suites

Or step by step:

make provision         # system deps (Debian)
make install           # pnpm install
make db-init           # create DB + role
make db-migrate        # Drizzle migrations
make db-seed           # demo enterprise, 5000 leads, 3 users, 2 pipelines

Bruno collection

cd collections/opentelecrm
bash ../../scripts/bruno-bootstrap-jwt.sh   # inject a dev JWT
npx @usebruno/cli run --env local -r .       # 27 requests, 58 assertions

Architecture

A Turborepo monorepo (pnpm@9.15.4, turbo run …). ESM-only, Node ≥ 22.

opentelecrm/
├── services/
│   ├── api/          # NestJS REST API  (Fastify, port 3005)
│   ├── mcp/          # JSON-RPC MCP     (Express, port 3006)
│   └── whatsapp/     # WhatsApp driver  (whatsapp-web.js, pairing-code)
├── packages/
│   └── db/           # Drizzle schema, RLS, migrations, seed data
├── apps/
│   ├── web/          # Next.js web app  (placeholder)
│   ├── widget/       # Embeddable widget(placeholder)
│   ├── extension/    # Browser extension(placeholder)
│   └── mobile/       # React Native app (placeholder)
├── collections/
│   └── opentelecrm/  # Bruno API test collection
├── docs/             # ARCHITECTURE.md, PARITY.md, ADRs
└── scripts/
    └── provision/    # Native Debian setup (no Docker)

Auth flow

  1. Dev JWT — HS256 signed with DEV_JWT_SECRET, used in local dev + Bruno
  2. API tokens — telekrm_async_ / telekrm_sync_ tokens, stored as SHA-256 hash
  3. OIDC — Zitadel id-tokens (future)

All requests route through AuthGuard + RLS via withTenant(eid) — no row can leak across tenants.

TeleCRM parity

Area Status
Foundation (multi-tenant, RLS, seed data) ✅
Lead CRUD + search + upsert ✅
Action logging (note, call, WhatsApp) ✅
Async autoupdate + validation ✅
API token management ✅
Team member list ✅
Custom actions ✅
WhatsApp (conversations, send, templates) ✅
Telephony (calls, dialer, callbacks, recordings) ✅
Web app, widget, extension, mobile 🚧 — API first, apps deferred

Full detail: docs/PARITY.md


Test surface

pnpm test              # all workspace tests
cd services/api && npx vitest run  # focused: 62 tests, 7 files
make typecheck         # tsc --noEmit across all workspaces
pnpm lint              # Biome

Tests spin up the real API (dev JWT + seeded DB), hit authGuard + RLS — no mocking.


Tech stack

  • Runtime — Node.js ≥ 22, ESM
  • Package management — pnpm + Turborepo
  • API — NestJS 10 + Fastify adapter
  • DB — PostgreSQL 16, Drizzle ORM + Drizzle Kit, native RLS
  • Auth — HS256 JWTs + SHA-256 API tokens + Zitadel OIDC (future)
  • Deps — TypeScript 5.7, Biome 1.9, Vitest
  • WhatsApp — whatsapp-web.js (wwebjs), pairing-code auth
  • Infra — Native Debian 13, systemd units, no Docker

License

AGPL-3.0 — see docs/LICENSES.md.

This is not affiliated with TeleCRM. It is an independent, clean-room implementation of the documented API surface for self-hosted use.

Recommended Servers

playwright-mcp

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.

Official
Featured
TypeScript
Magic Component Platform (MCP)

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.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

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.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

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.

Official
Featured
TypeScript
Kagi MCP Server

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.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

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.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured