magic: Build and Manage Enterprise AI Agents
Summary
Magic is a self-hostable platform for building AI agents and workflows and connecting them to organizational tools and knowledge. It combines collaboration, approval controls, sandboxing, and budget limits for teams that need managed AI automation.
At a glance
- Language
- TypeScript
- License
- NOASSERTION
- Stars
- 5k
- Forks
- 567
- Added to OSRepos
- October 31, 2025
- Last analyzed
- October 3, 2026
Topics
Click on any tag to explore related repositories
Use at your own risk
OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of code from these repositories is the user's own responsibility. Always review the repository, source code, dependencies, licenses, and security implications before running or installing anything. OSRepos is not responsible for issues, damages, or losses resulting from third-party repositories.
Overview
Magic is an AI agent platform that brings agents, workflows, messaging, and collaborative work into one system. It aims to help individuals and organizations turn AI work into usable deliverables and connect agents with internal systems, rather than relying on standalone chat tools.
It is most relevant when a team needs shared agents, controlled access to business data, human review of sensitive actions, or centrally managed AI usage. The repository provides a self-hosted deployment path as well as links to cloud services.
Key Features
- Build and use AI agents alongside workflow automation.
- Connect agents to internal systems such as ERP, CRM, and databases by encapsulating them as reusable digital employees.
- Render AI output as artifacts such as reports, dashboards, presentations, and spreadsheets.
- Run agents in sandbox containers with tenant isolation and a network proxy.
- Require human approval for high-risk actions.
- Set AI budgets by department, user, and agent.
- Collaborate on shared projects and report progress to WeCom, DingTalk, or Lark groups.
- Integrate with Anthropic Skills and OpenClaw Skills, as described by the project.
Use Cases
- A small team can use agents for recurring work such as customer support, marketing, or data analysis when it wants to extend capacity without adding separate tools for each task.
- An IT or operations team can expose internal systems and domain knowledge through reusable agents for staff across departments.
- A company handling sensitive data can evaluate self-hosting, sandbox isolation, and approval workflows to keep automation within managed boundaries.
- Finance or management teams can connect business data to agents to create reports and dashboards on demand.
Project Facts
- Language: TypeScript
- License: NOASSERTION
- Stars: 5k
- Forks: 567
- Topics: agent, agi, ai, gpt, llm, low-code, mcp, no-code, sandbox, super-magic, workflow
- Archived: No
Getting Started
The README lists Docker and curl as requirements for self-hosting on macOS or Linux. Run:
curl -fsSL https://getmagicrew.sh | bash
The deployment guide and further setup details are in the repository README.
Alternatives
- n8n: n8n focuses on visual app-to-app workflows with optional code, while Magic emphasizes managed AI agents with team collaboration, approvals, sandboxing, and budgets.
- nanobot: nanobot is a lightweight Python framework for personal agents and automation, while Magic targets team-managed agents and workflows connected to organizational tools.
Considerations
- Self-hosting requires Docker and curl; the README says Windows support is coming soon.
- The README describes a broad enterprise platform, so teams should assess the deployment and integration work needed for their environment.
- The repository reports its license as
NOASSERTION. The README says the project uses a Magic Open Source License based on Apache 2.0 with additional restrictions, so review the license file before adoption.
Source repository
Open the original repository on GitHub.
16 counted GitHub visits
Related repositories
Similar repositories that may be relevant next.
OrcaReplay: Time Travel for AI Agents, Debugging and Evaluation
October 2, 2026
OrcaReplay introduces "time travel" capabilities for AI agents, allowing developers to record, replay, fork, and debug any agent run with any model. It addresses the challenges of AI agent debugging by providing byte-for-byte reproducibility, offline analysis, and the ability to compare different models from specific checkpoints. This tool, built by the OrcaRouter.ai team, enhances observability and control over complex agent behaviors.
OpenMake LLM: Self-Hosted AI Workspace for Local and Open-Weight LLMs
October 1, 2026
OpenMake LLM is an open-source, self-hosted AI workspace for local and open-weight LLMs. It coordinates specialized models, autonomous agents, and tools for deep research and artifact generation. This platform supports vLLM, LiteLLM, and BYOK providers, offering a robust environment for managing AI workloads.
ZenNotes: Keyboard-First Markdown Notes with Vim, Diagrams, and MCP Integration
October 1, 2026
ZenNotes is a versatile, keyboard-first Markdown notes app designed for speed and flexibility. It stores notes as plain Markdown files, offering Vim-friendly editing, diagram support, and integration with MCP tools. Available as a desktop app (Electron) and a self-hosted web app, ZenNotes provides a powerful solution for organizing your thoughts.
lat.md: A Knowledge Graph for Your Codebase, Written in Markdown
September 26, 2026
lat.md is an innovative tool that transforms your codebase knowledge into an interconnected graph of markdown files. It helps both AI agents and human developers quickly understand project architecture, business logic, and design decisions. By integrating directly into your project, lat.md ensures documentation remains consistent and up-to-date.