Local Article Studio

This repo now contains:

• backend/: FastAPI + CrewAI + SQLite
• frontend/: React + Vite + Tailwind + React Quill
• content_team.py: a thin CLI entrypoint into the new backend article generator

Start The Backend

cd C:\Users\milli\content_writing_ai_agents\content_writers
python -m venv .venv
.venv\Scripts\activate
pip install -r backend/requirements.txt
copy .env.example .env
python run_backend.py

Alternative, if you prefer the raw Uvicorn command, run it from the repo root only:

cd C:\Users\milli\content_writing_ai_agents\content_writers
.\.venv\Scripts\python.exe -m uvicorn app.main:app --reload --app-dir backend

Start The Frontend

cd frontend
npm install
npm run dev

Environment Notes

• Set LLM_PROVIDER, LLM_BASE_URL, and LLM_MODEL in .env to match your local Ollama or LM Studio setup.
• The default model in .env.example is qwen2.5:32b because the previous llama3.1 reference was not available in the local Ollama list discovered during debugging.
• Saved articles go to SQLite at backend/data/articles.db.
• CrewAI runtime storage is redirected into backend/.cache/crewai.
• If you previously upgraded duckduckgo-search manually, rerun pip install -r backend/requirements.txt so the local environment matches the project requirements.