# Zetta CRM

> Open-source MCP backend for AI agents

**URL:** https://www.ciptadusa.com/incubating/zetta-crm  
**Stage:** beta  
**Published:** 2026-06-15  
**Repository:** https://github.com/incredible-zetta/crm  
**Live product:** https://zettacrm.com/  
**Docs:** https://zettacrm.com/  

## Product

Go-based MCP server with 34 complete agent tools: contacts, email, campaigns, tracking links, scheduling, analytics, exports, unsubscribe flow, soft-delete, purge, and IMAP inbox replies. Single binary, self-hosted, MySQL backend, SMTP/Mailgun email providers. v0.4.0 stable release — GHCR image ready to deploy for production pilots.

API-key gated, single database, POST /mcp endpoint. Built for AI agents that need a self-contained CRM backend without cloud vendor lock-in.

Suited for: solo builders, small team sales ops, AI agent workflows, and anyone who wants a CRM backend that can be audited and fully controlled.

## Media

- image: https://cdn-uagents.enitip.com/uploads/products/zetta-crm/screenshot.png — Zetta CRM live landing page (zettacrm.com) — CRM backend built for AI agents.
- image: https://cdn-uagents.enitip.com/uploads/products/zetta-crm/showcase-inbox-analytics.jpg — Real usage showcase: user asks 'Cek statistik di zetta dan cek inbox' via WhatsApp, bot reports analytics + inbox summary, and self-detects an open_rate anomaly worth flagging to dev (7 June 2026, 13:53 UTC).

## Updates

### v0.5.0 — Threads hybrid channel, topic tags, reply tree

**Published:** 2026-06-15

v0.5.0 just shipped to GHCR. Major milestone: **Threads** is now the first hybrid channel beyond Email & WhatsApp. 34 → 38 MCP tools total.

**What's new:**

- 🆕 **Threads hybrid channel** — publish, reply, reply_tree, search, history, profile, token management, reply quota. End-to-end via MCP.
- 🆕 **Topic tags** — full support for Meta's official Threads topic tag system (1-50 char, no `.`/`&`).
- 🆕 **threads_reply_tree** — nested reply conversation with `is_mine` + `needs_reply` flags. Makes reply-out automation straightforward.
- 🆕 **Expanded Threads search** — TOP/RECENT, KEYWORD/TAG, media type, author username, time range.
- 🐛 **Fixed threads_reply_tree schema cycle** — flattened output so MCP schema is valid.
- 🐛 **Fixed threads_history serialization** — robust for pagination.
- 🔧 **CI lint + docs cleanup** for Threads integration.

**Strategy:** Zetta CRM now supports 3-channel outbound (Email + WhatsApp + Threads) through one backend. Ideal for AI agent workflows that need omnipresence across social + messaging.

Repo: github.com/incredible-zetta/crm/releases/tag/v0.5.0
Install: `docker pull ghcr.io/incredible-zetta/crm:v0.5.0`

### Moved to beta

**Published:** 2026-06-15  
**Stage change:** beta

v0.5.0 ships Threads hybrid channel. 38 MCP tools total (34 → 38). 3-channel outbound now production-ready: Email + WhatsApp + Threads. Production pilots expanding.

### Moved to incubating stage

**Published:** 2026-06-15

After v0.1.0 stable (May 2026) and 4 minor releases (v0.1.2 → v0.4.0), Zetta CRM officially entered the **incubating** stage in the CDS product portfolio.

Status: stable API, multi-platform deployment ready, 34 MCP tools live, complete wiki documentation. Next target: 10 active production pilots and movement to **beta** stage in Q3 2026.

Repo: github.com/incredible-zetta/crm
Live: zettacrm.com
Install: docker pull ghcr.io/incredible-zetta/crm:v0.4.0

### v0.4.0 released — production-ready GHCR image

**Published:** 2026-06-10

v0.4.0 released on GHCR (`ghcr.io/incredible-zetta/crm:v0.4.0`). Multi-platform image, install wiki live, release notes published. Cloud install guides complete: EasyPanel, Coolify, Railway, Render, Fly.io, AWS ECS, Google Cloud Run, Kubernetes.

Current status: 4 stars on GitHub, early production pilots on-going. Stage: incubating — moving toward beta with focus on observability + multi-tenant auth.

Next milestone: v0.5.0 with built-in Web UI dashboard, audit log, and webhook subscriptions.

---

*Markdown version of https://www.ciptadusa.com/incubating/zetta-crm — generated for AI agents and LLM crawlers.*