SparkCrew

사람과 AI가 같은 맥락에서 대화하고, 작업하고, 결과를 만드는 AI 협업 프로젝트
An AI collaboration project where people and AI share context, work, and produce results together

개인 AI 대화와 팀 Topic/Thread 협업을 바탕으로 파일·지식·백그라운드 작업과 Browser/Terminal/Workspace 실행을 연결하는 방향입니다.
See the Korean and English documentation for details.

한국어  ·  English

---

Project direction

SparkCrew explores a collaboration model in which conversation is shared context rather than the only workspace.

• Personal AI: private AI conversations and personal topics.
• Team topics and threads: SNS-style posts and threaded discussion for people and shared AI participants.
• Files and knowledge: shared files remain separate from RAG indexing scope; upload does not automatically promote a file to team or organization knowledge.
• Background work: AI tasks run independently from the conversation that requested them.
• Execution tools: browser automation, terminal tasks, and isolated workspaces can be attached to a task when needed.
• Shared results: documents, images, video, charts, tables, notebook/HTML results, and live browser sessions can be presented in a shared viewing surface.

Full desktop/OS streaming and control are not part of the current project direction. Browser-based Computer Use is the primary interactive execution target.

Current runnable scaffold

• Frontend: Next.js user UI at /, a separate product Console at /console/*, React, TypeScript, Tailwind CSS, axios, SweetAlert2, and Node Playwright tests.
• Backend: Django 6 on Python 3.12–3.14, using one Django project (config) with two Django apps (core, agent).
• URLs: core DRF at /core/*, Agent FastAPI at /agent/*, and Django Admin at /admin/*.
• Composition: config.asgi.application mounts FastAPI and Django into one ASGI application, served identically by Daphne-backed manage.py runserver or Uvicorn.
• Browser foundation: backend Python Playwright provides the async Agent Browser Computer Use package boundary; it is separate from frontend Playwright tests.

The scaffold implements health endpoints and package boundaries only. It does not implement domain models, RAG, LLM orchestration, browser sessions, Terminal, or Workspace behavior.

Local development

python -m pip install -r backend/requirements.txt
playwright install chromium
python backend/manage.py check
python backend/manage.py test core agent
python backend/manage.py runserver 127.0.0.1:8000

cd frontend
npm install
npm run lint
npm run build
npm run test:visual
npm run dev -- --hostname 127.0.0.1 --port 3000

Open http://127.0.0.1:3000 for the user surface and http://127.0.0.1:3000/console for the product Console. Django Admin remains at http://127.0.0.1:8000/admin/.

Optional Django Admin setup

The product Console at /console/* and Django's ORM-backed internal Admin at /admin/* are separate UIs. Initialize the built-in Django tables and create a local administrator before signing in to Django Admin:

python backend/manage.py migrate
python backend/manage.py createsuperuser
python backend/manage.py runserver 127.0.0.1:8000

backend/db.sqlite3 is a local development artifact and must not be committed. The scaffold includes no custom domain migrations.

---

Language

Language	README
한국어	docs/README.ko.md
English	docs/README.en.md

---

License

Source code is licensed under the PolyForm Noncommercial License 1.0.0. See the repository license file for details.