magic  by dtyq

AI productivity platform for building/deploying AI apps

Created 4 months ago
4,190 stars

Top 11.8% on SourcePulse

GitHubView on GitHub
Project Summary

Magic is an open-source, all-in-one AI productivity platform designed to help enterprises build and deploy AI applications for a 100x productivity increase. It offers a suite of integrated products including a general-purpose AI agent (Super Magic), an AI-powered instant messaging system (Magic IM), and a visual AI workflow orchestrator (Magic Flow), targeting businesses seeking to leverage AI for complex tasks, internal communication, and process automation.

How It Works

Magic operates as a product matrix, integrating several AI-centric tools. Super Magic utilizes a multi-agent design system and tool capabilities for autonomous task understanding, planning, action, and error correction. Magic Flow provides a visual, node-based interface for orchestrating AI workflows, supporting any LLM compatible with the OpenAI API. Magic IM focuses on enterprise communication and knowledge management, featuring knowledge base retrieval and group chat integration with AI participation.

Quick Start & Requirements

  • Install: Clone the repository and use the provided script: git clone https://github.com/dtyq/magic.git, then ./bin/magic.sh start.
  • Prerequisites: Docker 24.0+, Docker Compose 2.0+. Requires LLM environment variables to be configured in .env and env_super_magic.
  • Access: API at http://localhost:9501, Web App at http://localhost:8080.
  • Docs: https://docs.letsmagic.cn/en

Highlighted Details

  • Super Magic is designed for complex task scenarios with autonomous capabilities.
  • Magic Flow offers visual orchestration and supports any OpenAI-compatible LLM.
  • Magic IM provides enterprise-grade messaging with knowledge base integration and multi-organization support.
  • Future components include Teamshare OS, Magic Use, Magic Space, and Sandbox OS.

Maintenance & Community

The project encourages contributions and provides a contribution guide. Security vulnerabilities can be reported via email to dev@dtyq.com.

Licensing & Compatibility

The project uses the "Magic Open Source License," described as essentially Apache 2.0 with additional restrictions. Compatibility for commercial use or closed-source linking would require careful review of these restrictions.

Limitations & Caveats

Several key components (Teamshare OS, Magic Use, Magic Space, Sandbox OS) are marked as "Coming soon," indicating the platform is still under active development. Access to cloud services currently requires an invitation code.

Health Check
Last Commit

1 week ago

Responsiveness

1 day

Pull Requests (30d)
1
Issues (30d)
9
Star History
1,327 stars in the last 30 days

Explore Similar Projects

Starred by Andrew Ng Andrew Ng(Founder of DeepLearning.AI; Cofounder of Coursera; Professor at Stanford), Thomas Wolf Thomas Wolf(Cofounder of Hugging Face), and
4 more.

ag2 by ag2ai

0.9%
4k
AgentOS for building AI agents and facilitating multi-agent cooperation
Created 10 months ago
Updated 2 days ago
Starred by Paul Stamatiou Paul Stamatiou(Cofounder of Limitless), Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"), and
5 more.

claude-task-master by eyaltoledano

1.0%
22k
AI-powered task management system for code editors
Created 6 months ago
Updated 1 day ago
Feedback? Help us improve.