Structured task setup
Objective, output type, platform, constraints, model lane, and execution plan keep AI work repeatable.
Structured tasks, drafts, reviews, and scheduling — all local-first.
Plan, draft, review, and schedule without losing control of the process.
Objective, output type, platform, constraints, model lane, and execution plan keep AI work repeatable.
Generate variants, review, edit, approve, and schedule — no live posting without explicit approval.
Move tasks through idea, drafting, review, approval, and scheduling with drag-and-drop cards.
Optional Ollama integration with local fallback so the app works even when no model server is running.
Run the backend and frontend in separate terminals, or use the provided Docker Compose stack.
git clone https://github.com/LumenHelixLab/socialforge-ai.git cd socialforge-ai # Terminal 1: backend cd backend && python3 -m venv .venv && .venv/bin/pip install -r requirements.txt && .venv/bin/uvicorn main:app --reload --port 8787 # Terminal 2: frontend cd frontend && npm install && npm run dev
Tested on Windows 11, macOS Sonoma, Ubuntu 22.04/24.04, and modern mobile browsers.
A FastAPI + SQLite backend serves a React + Vite job board and inspector.
Creator -> React/Vite Frontend -> FastAPI Backend -> SQLite + Local Model Adapter
| |
└────────── Drag-and-drop job board + calendar + inspector ────┘Everything you need to run, extend, and deploy socialforge-ai.