Wenlan
Wenlan은 소스 기반 AI 지식베이스·LLM 위키로, 문서·결정을 페이지화해 검색·검토·리뷰 가능
작업을 수행하는 AI 동료, 자체 호스팅 가능, AGPL‑3.0
AI coworkers that run your work. Self-hostable, AGPL-3.0.
생산성GitHub @rerun-build직접 설치운영자 미확인
써 보셨어요? 한 번만 눌러 주세요. 로그인은 필요 없어요.
저장소 정보는 아직 수집하지 않았습니다.
아직 감지된 업데이트가 없습니다. GitHub 릴리스가 나오면 여기에 보입니다.
메이커가 쓴 소개는 위의 한 줄이 전부입니다. 더 자세한 내용은 저장소 README에서 읽을 수 있습니다. 저장소 README ↗
README 전체를 불러오는 중입니다.
README 전문 보기 ↗You tell Rerun what you need done, in plain words. The Chief of Staff interviews you, plans a team and builds it. Each coworker gets a job, its own memory, its own skills and its own computer. You see them all on one board, you talk to any of them like to a colleague, and they ask before they send, pay or delete anything.
A few coworkers people hire on day one, each a ready-made template:
🧑✈️ The Chief of Staff runs your team. Say what you need, out loud if you like. It spots what needs doing, hires the coworker for it, and answers your coworkers by your rules while you are away.
And everything else a colleague needs:
Let an AI install it for you:
Or paste this into Claude Code, Codex or Cursor, which run the commands for you:
Install Rerun, the open source AI coworkers app, for me. Follow the guide written for AI assistants: https://raw.githubusercontent.com/rerun-build/rerun/main/INSTALL.md
By hand, you need Docker, Node.js 22, and a Supabase project (a free one on supabase.com works).
git clone https://github.com/rerun-build/rerun.git && cd rerun
cp .env.example .env # fill the Supabase URL and keys, and an OpenRouter key
npx supabase link --project-ref <your-project-ref> && npx supabase db push --include-seed
node scripts/self-host-check.mjs
docker compose up -d --build # then open http://localhost:3000
That's it, say hi to the Chief of Staff. 🎉
The self-hosting guide walks through every step, including Supabase on your own machine, adding servers, HTTPS and updates.
Same product, same code. Rerun Cloud is this Rerun, hosted, plus what only a hosted service can offer.
| 🛠️ Open source | ☁️ Rerun Cloud | |
|---|---|---|
| Coworkers, chat, memory, skills, schedules | ✅ | ✅ |
| Services catalog, MCP, Composio | ✅ | ✅ |
| Template library | Your own templates | ✅ |
| A computer for each workspace | You host it, one command attaches it | Ready in 30 seconds |
| AI models | Your keys or subscriptions | Included credit, or your keys |
| Premium features (LinkedIn, WhatsApp, paid data, emails) | ✅ | |
| Updates | Each release | Continuous |
| Support | Community | ✅ |
| Price | Free forever | See plans |
Part of Rerun Cloud runs on two partners: LinkedIn, WhatsApp and paid data. Their integrations live in this repository, in the open.
Need a coworker built around your business, or a hand to set Rerun up? Rerun Experts are vetted builders, freelancers and studios who design, ship and run AI coworkers on Rerun.
flowchart LR
you(["🧑 You"]) --> app["<b>App</b><br/>Next.js · apps/ui"]
app <--> db[("<b>Supabase</b><br/>Postgres, auth,<br/>storage, realtime")]
app <--> engine["<b>Engine</b> · apps/engine<br/>one per workspace"]
engine --> coworkers["🤖 Coworkers<br/>files, browser, screen"]
engine <--> tools["🔌 Your tools<br/>MCP, APIs"]
engine <--> models["🧠 Your models"]
Built with Next.js, React, Tailwind, Supabase, the Vercel AI SDK, Hono, libSQL, Patchright and Docker, in a pnpm monorepo.
Bug reports, fixes and docs are welcome. Read CONTRIBUTING.md first: it explains how to run Rerun from source and how pull requests flow back into Rerun Cloud.
Self-hosted installs get community support on GitHub: an issue without logs or steps to reproduce is closed with a link to the guide. Questions and show-and-tell go to Discord.
If Rerun saves you a few hours, a ⭐ helps a lot.
AGPL-3.0. You can run, modify and host Rerun freely; if you host it for others, you publish your changes under the same license. The name and logo are not covered: see TRADEMARK.md.
Directories named ee or (ee) (billing and plans, the Rerun credit and its managed models, the
template marketplace) are source-available under the Rerun Commercial License:
read and modify them freely, run them in production with an agreement. The open source build never
switches them on.