magic
Reduces development cycle time by automating routine coding tasks and scaffolding, allowing engineering teams to focus on architecture and complex problem-so…
Magic Cloud is an open-source, self-hostable AI software development platform that generates and runs full-stack business applications and AI agents from natural language. Built on OpenAI and Hyperlambda (a backend DSL), it includes a CRUD generator, SQL Studio, a web-based IDE, task scheduler, and built-in web server. It runs as a standalone Docker deployment with its own dashboard — not a Claude Code skill — though an optional 'mcp' plugin can expose your Magic endpoints to Claude Code as an MCP server.
- Generate complete API endpoints from natural language specifications and requirements
- Automatically create database schemas and migrations based on application data models
- Build full-stack features including backend logic, frontend components, and database setup
Reduces development cycle time by automating routine coding tasks and scaffolding, allowing engineering teams to focus on architecture and complex problem-solving rather than boilerplate implementation.
Development teams seeking to accelerate delivery velocity on greenfield projects or standardized feature implementations.
https://github.com/polterguy/magic
By polterguy
How to Get It
Create the docker-compose.yaml from the README (servergardens/magic-backend image), run 'docker compose up', then visit localhost:5555 (login root/root)
Tip: Paste this into a Claude Code conversation. Verify command matches your Claude Code version.
After installing, paste this into Claude:
Help me generate complete API endpoints from natural language specifications and requirements
Trust Signals Auto-scanned
Community Pulse Growing
Discussed on Hacker News
- Functioning Open Source Version of Lovable/Bolt/Replit — Hacker News · 2 pts
- Magic: A low-code CRUD generator — Hacker News · 2 pts
- Magic — Hacker News · 2 pts
3 mentions across 1 sources
Reviewer notes
Auto-scanned review. These are observations, not a security certification.
Scored from trust signals (evidence-eval-v1): 1,146 GitHub stars; contributors unknown; last commit 0d ago; license MIT.
Things to check
- Standalone self-hosted platform, not a Claude Code skill. Depends on OpenAI: you need an API key, and the default model may require a funded OpenAI account; embeddings are OpenAI-only even if you swap inference to Ollama or HuggingFace.
How to evaluate tools before deploying →
Data shown here comes from public APIs and automated scanning. Reviewer notes reflect one person's experience. This is not a security certification or legal recommendation. Always evaluate tools according to your own organization's policies.
Evaluation
Scored from trust signals (evidence-eval-v1): 1,146 GitHub stars; contributors unknown; last commit 0d ago; license MIT.