aozorae/Edgechat

A modern, self-hosted team chat built on Cloudflare — realtime messaging, Telegram bridge, encrypted messages & files, and WebMCP. | 基于 Cloudflare 的现代自部署团队聊天系统：实时通信、Telegram 双向桥接、消息与文件加密，以及 WebMCP。

What it solves

EdgeChat provides a way for teams and small groups to have their own independent, self-hosted chat space without the burden of maintaining a dedicated server or database. It eliminates the need for permanent server infrastructure by leveraging serverless edge computing.

How it works

The system is built entirely on the Cloudflare ecosystem. It uses Cloudflare Workers for the backend logic, Durable Objects and WebSockets for real-time communication, D1 for the database, KV for session storage, and R2 for file storage. Deployment and updates are automated via GitHub Actions.

Who it’s for

Developers, small teams, and communities who want a private, self-deployed chat system that is easy to maintain and integrates with existing Telegram groups.

Highlights

  • Telegram Bridge: Bi-directional message forwarding between EdgeChat groups and Telegram groups via a bot.
  • Serverless Deployment: No permanent server required; runs on Cloudflare Workers.
  • Server-side Encryption: Uses AES-256-GCM for new message bodies and attachments.
  • Comprehensive Chat Features: Supports public/private groups, direct messages, voice messages, and file sharing.
  • Multi-platform Access: Includes a responsive web interface and a Capacitor-based Android app.

Related

  • Project
    Accio-Lab/Dressage

    A scalable reinforcement learning framework for LLM agents that enables training across diverse agent types and sandboxes with token-level trajectory recording.

  • Project
    openai/openai-cli

    The official OpenAI CLI for interacting with the OpenAI REST API, enabling resource management and API calls directly from the terminal.

  • Project
    chainbase-labs/manuscript-core

    A protocol and framework for simplifying omnichain data access and processing, enabling developers to unify blockchain data from multiple sources into a a single pipeline.

  • Project
    enoche/MMRec

    A modern multimodal recommendation toolbox that provides a unified framework and source code for a wide variety of research-based recommendation models.