windinwing

windinwing / mchat

Public

MChat is a lightweight, embeddable, multi-tenant AI customer service platform. It integrates a powerful Bot Engine, RAG knowledge base, Skill plugin system, and an embedded chat Widget — with support for 10+ LLM providers and multi-channel connectivity (Web Widget, WebSocket, REST API, and extensible channel adapters for messaging apps).

12
0
100% credibility
Found May 25, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

MChat is an open-source multi-tenant AI customer service platform that lets businesses create intelligent chatbots powered by large language models, with built-in knowledge base search, skill plugins, and embeddable chat widgets for websites.

How It Works

1
💬 You want to add AI support to your website

You discover MChat and learn it can create a smart customer service chatbot that answers questions from your knowledge base.

2
🚀 You launch the platform with one click

Using Docker, you start the full system including the admin panel, AI backend, and frontend in minutes.

3
🔑 You connect your AI service

You enter your AI provider details (like OpenAI or Claude) so the chatbot can think and respond intelligently.

4
📚 You teach the chatbot your business

You upload your documents, manuals, or FAQs so the chatbot knows about your products and services.

5
🎨 You customize how it looks and works

You choose colors, welcome messages, and which skills the chatbot should have for your visitors.

6
📋 You get a small piece of code to add

The platform gives you a simple script tag that you paste into your website to show the chat window.

🎉 Your customers can now chat with your AI assistant

Visitors to your website see your branded chat widget, ask questions, and get instant answers from your knowledge base.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 12 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is mchat?

MChat is a lightweight, embeddable multi-tenant AI customer service platform built with Python and React. It gives you a streaming bot engine, RAG knowledge base, skill plugin system, and embeddable chat widget all in one package. You can connect to 10+ LLM providers and deploy across multiple channels including Web Widget, WebSocket, REST API, and WeChat with a single docker compose command.

Why is it gaining traction?

The hook is the embeddable widget: one script tag gives you a full branded chat interface on any website. The multi-tenant architecture means you can run separate AI configurations for different customers without managing multiple deployments. For teams running Ollama locally, this gives you a complete support platform without sending data to external APIs.

Who should use this?

Small teams building SaaS products who want to add AI customer support without locking into third-party services. Developers already running local models who need a full-stack solution with conversation management and knowledge base features. Startups that need to embed chat on multiple client websites while keeping data isolated per customer.

Verdict

MChat solves real problems and has a coherent architecture, but the 1.0% credibility score reflects its early stage. With only 12 stars and limited community feedback, you are essentially an early adopter. The Docker setup works and the API is documented via Swagger, but test coverage and documentation depth are still catching up. Worth exploring if you want to self-host AI support, but budget time for the inevitable edge cases you will encounter.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.