I built MuseBot, a multi-platform chatbot that integrates with LLM APIs to provide AI-powered responses. It supports platforms like Telegram, Discord, Slack, Lark (Feishu), DingTalk, WeCom (企业微信), QQ, and WeChat, so you can talk to your favorite model anywhere.

MuseBot connects seamlessly with OpenAI, DeepSeek, Gemini, and OpenRouter models to make conversations feel natural, dynamic, and responsive.

Key Features

AI Responses – Intelligent chatbot replies using LLM APIs.

Streaming Output – Real-time responses that feel conversational.

Easy Deployment – Run locally or on any cloud server in just a few steps.

Image Understanding – Send an image, and the bot can interpret and respond.

Voice Support – Communicate using voice messages.

Function Calling – Supports MCP-style function calls for extending capabilities.

RAG Support – Enhances context understanding with retrieval-augmented generation.

Admin Platform – Web-based platform to manage bots and configurations.

Service Registration – Automatically register bot instances to a service registry.

Metrics and Monitoring – Built-in Prometheus metrics for observability.

MuseBot is built entirely in Golang, designed for performance, modularity, and easy extensibility. I’d love to hear feedback from developers working with chatbots, LLM integrations, or Go-based infrastructure — especially ideas to improve scalability and real-time performance.