PromptFlow is an open-source, local-first AI Agent Builder Platform that allows users to visually create and execute custom AI workflows through a drag-and-drop canvas. It supports prompt engineering, retrieval-augmented generation (RAG), tool calling, and real-time streaming.
- Visual Canvas Editor – Drag and drop nodes to build AI workflows
- RAG Support – Upload documents and retrieve context for prompts
- Local-First – Everything runs locally with open-source components
- Real-time Streaming – Watch your agents execute live
- Tool Integration – Extend agents with custom tools
- Team Collaboration – Multi-user workspaces with role-based access
- Frontend: Next.js 15 + TypeScript + Tailwind CSS + React Flow
- Backend: FastAPI + Python 3.11
- Databases: PostgreSQL + Qdrant (vectors)
- Storage: MinIO (S3-compatible)
- Cache/Queue: Redis
- LLM Runtime: Ollama
- Task Queue: Celery
- Docker and Docker Compose
- Node.js 18+
- Python 3.11+
- Ollama (for local LLMs)
- Clone the repository
git clone <https://github.com/prajjwal-23/PromptFlow.git>
cd PromptFlow- Start all services
make setup
make dev- Access the application
- Frontend: http://localhost:3000
- Backend API: http://localhost:8000
- API Docs: http://localhost:8000/docs
# Install dependencies
make install
# Start development servers
make dev
# Run tests
make test
# Lint code
make lintPromptFlow/
├── backend/ # FastAPI backend
├── frontend/ # Next.js frontend
├── docker-compose.yml
├── Makefile
└── docs/ # Documentation
Please read CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.
This project is licensed under the MIT License - see the LICENSE file for details.
- Built with FastAPI
- UI powered by Next.js and React Flow
- Vector search with Qdrant
- LLMs via Ollama