Deco Chat is an open-source, self-hosted AI workspace designed for developers to rapidly build, prototype, and deploy intelligent, UI-rich AI agents and applications that integrate with internal systems.

Who is Deco Chat designed for?

It's ideal for 'Vibecoders' prototyping new AI ideas and 'Agentic engineers' focused on deploying scalable, secure, and sustainable AI-powered production systems.

What are Deco Chat's core technical capabilities?

Key features include an open-source runtime, MCP Mesh for secure model/data/API integrations, a unified TypeScript stack (React/Tailwind, typed RPC), and global Cloudflare-based infrastructure for scalable self-hosted deployments.

How does Deco Chat facilitate AI agent development?

It provides 'Tools' for atomic functions (calling APIs, databases, AI models) and 'Workflows' to orchestrate these tools using Mastra operators, enabling seamless composition of complex AI logic into agents.

Deco Chat

Name: Deco Chat
Author: deco-cx

bydeco-cx

•

198

•

开发者工具

数据科学与机器学习

云基础设施

Provides an open-source, self-hosted AI workspace for developing and deploying intelligent, UI-rich AI agents that integrate with internal systems.

Deco Chat

bydeco-cx

•

198

•

开发者工具

数据科学与机器学习

云基础设施

Provides an open-source, self-hosted AI workspace for developing and deploying intelligent, UI-rich AI agents that integrate with internal systems.

Deco Chat serves as an open-source foundation for building AI-native software, equipping developers, engineers, and AI enthusiasts with robust tools to rapidly prototype, develop, and deploy AI-powered applications. Its core mission is to enable teams to embrace Generative AI by providing the necessary building blocks for creating sustainable, governed AI-native applications, scalable from initial prototype to large-scale production environments and thousands of users.

主要功能

01Open-source runtime to compose tools, workflows, and views within a single codebase.

02MCP Mesh for securely integrating models, data sources, and APIs with observability and cost control.

03Unified TypeScript stack combining backend logic and custom React/Tailwind frontends via typed RPC.

04Global, modular infrastructure built on Cloudflare for low-latency, infinitely scalable self-hosted deployments.

05Visual workspace to build agents, connect tools, manage permissions, and orchestrate code-defined elements.

0676 GitHub stars

使用案例

01Rapidly prototyping new AI-powered application ideas.

02Developing and deploying scalable, secure, and sustainable production AI systems.

03Integrating AI agents seamlessly with internal tools, data sources, and existing APIs.