#99th of 17 in Agent platforms

openmaic

Generates interactive classes with AI teachers and classmates from a topic or file

Stars
40.7k
License
MIT
Last commit
Oct 2026
Last release
Oct 2026
Language
TypeScript
Stars, last 29 weeks+29,361
25 Mar 2026: 11,378 stars1 Apr 2026: 12,754 stars8 Apr 2026: 14,053 stars15 Apr 2026: 15,155 stars22 Apr 2026: 15,745 stars29 Apr 2026: 16,058 stars6 May 2026: 16,334 stars13 May 2026: 17,030 stars20 May 2026: 17,359 stars27 May 2026: 17,636 stars3 Jun 2026: 17,904 stars10 Jun 2026: 18,120 stars17 Jun 2026: 18,318 stars24 Jun 2026: 18,492 stars1 Jul 2026: 18,887 stars8 Jul 2026: 19,146 stars15 Jul 2026: 19,414 stars22 Jul 2026: 19,706 stars29 Jul 2026: 20,025 stars5 Aug 2026: 20,330 stars12 Aug 2026: 20,489 stars19 Aug 2026: 20,646 stars26 Aug 2026: 20,792 stars2 Sept 2026: 30,496 stars9 Sept 2026: 34,524 stars16 Sept 2026: 37,312 stars23 Sept 2026: 38,705 stars30 Sept 2026: 39,614 stars7 Oct 2026: 40,094 stars14 Oct 2026: 40,739 stars

11,378 on 25 Mar 202640,739 on 14 Oct 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

OpenMAIC turns a topic or attached documents into a lesson made of slides, quizzes, interactive HTML simulations and project-based learning activities. AI teacher and classmate agents speak via TTS, draw on a whiteboard and discuss with the learner. Built on Next.js 16, React 19 and LangGraph. Since 1.2.0, course generation runs on the server and courses are stored in PostgreSQL.

Who it is for: Educators and self-hosters building AI-taught courses from documents

Strengths

  • Exports lessons as editable .pptx or interactive .html pages
  • Many providers: OpenAI, Anthropic, Bedrock, Gemini, Ollama, Lemonade, any OpenAI-compatible API
  • Models set once per slot in openmaic.yml, with lock and allowUserKeys: false options
  • OpenMAIC Skill lets OpenClaw and similar workbenches generate classrooms from chat apps

Weaknesses

  • 1.2.0 needs PostgreSQL 16 and a long-running server; Vercel and serverless are unsupported
  • Latest 1.2.0 line is a release candidate; changelog must be read before upgrading
  • README recommends image, video, TTS and web search keys for full function
  • Compose install is single-user; shared access needs ACCESS_CODE and manual network publishing

What it needs

  • GPU optional
  • Docker + Compose
  • Needs PostgreSQL 16, Node.js >= 22.19, pnpm >= 10, LLM provider API key
  • Models: OpenAI, Anthropic, Amazon Bedrock, Google Gemini, DeepSeek
  • port 3000

Also in Agent platforms

See all 17
Also in Agent platforms
RankProjectScore
1LangflowVisual builder for AI agents and workflows, exposed as API or MCP server155.7k starsMIT80 out of 100
2daily_stock_analysisDaily LLM stock analysis for A-share, HK, US and other markets, pushed to chat apps66.4k starsMIT77 out of 100
3deer-flowAgent harness that runs sub-agents, memory and sandboxed skills for long tasks83.9k starsMIT75 out of 100
#4DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs Live demo ↗ (opens in a new tab)158.5k starscustom license74 out of 100
#5MulticaIssue board where AI coding agents take assignments like teammates52.8k starscustom license72 out of 100