Chengos

ChengOS is a visual AI data processing and agent workflow system built around a Rust DAG engine. Users design workflows in a visual editor, run ReAct agents that discover and call external tools, and connect those agents to chat channels, all in one self hosted platform. It follows a client server design with a REST API and WebSocket streams, and ships four components: the backend engine, a React management console, an app and chat gateway, and a terminal client. It is aimed at developers and teams that want to orchestrate AI agents and data pipelines on their own infrastructure. A built in MCP client supports Stdio, SSE and HTTP transports. More than 60 built in node tools are available, and skills can be generated from natural language, imported from a URL or Git repository, or loaded as standard SKILL.md packages with validation and hot reload. Credentials sit in an encrypted vault that models never see, alongside sandboxing, multi tenant role based access control, vector retrieval and scheduling. A local Word and Excel canvas lets agents process documents without external office platforms. The chat gateway binds apps to Telegram, WhatsApp, Slack, WeCom and DingTalk, or to an embedded web widget and mobile friendly chat, and the terminal client offers agent chat in the style of Claude Code or Codex. The site states that ChengOS is free for personal and commercial use, with an enterprise edition announced as coming soon. The pricing page lists Free, Pro and Business subscriptions covering one, three and ten platforms, with Pro adding data export and API access and Business adding cloud sync, priority support and advanced analytics, but no amounts are shown. It installs as a native binary or with Docker, and the source is on GitHub.