Tired of manual API testing?. API Flow lets you automate complex workflows using plain English.
Before API Flow (The Pain):
- Open Postman.
- Build
POST /auth/login. - Run it, then frantically copy the auth token.
- Build
GET /user/profile. - Paste the token into the header.
- Run it.
- Repeat until you question your life choices.
With API Flow:
"Log in with username 'test' and password 'pass', then get my user profile"
That's it. π
Watch the demo to see API Flow in action: πΉ Demo Video
An AI-powered engine (using Google Gemini & LangGraph) acts as your brilliant, tireless intern. It takes your request, creates a multi-step plan, calls the APIs, and even cleverly passes data (like auth tokens) between steps. All you see is the result.

Tech Stack:
- Backend: FastAPI, LangGraph, Python
- Frontend: Next.js, React, TypeScript
Prerequisites:
- Python 3.11+
- Node.js 18+
- A Google Gemini API Key
Setup:
1. Clone the Repository
Bash
git clone https://github.com/usman-faisal/api-flow.git
cd api-flow
2. Backend Setup & Run
cd backend
# Install dependencies with uv (recommended) or pip
uv sync
# OR
pip install -r requirements.txt
# Create environment file
cp .env.example .env
# Add your GEMINI_API_KEY to .env
# Start the FastAPI server
uv run fastapi dev
# OR
python -m uvicorn app.main:app --reload --port 8000
3. Frontend Setup & Run
cd frontend
# Create environment file
cp .env.example .env
# Install dependencies
pnpm install
# Start the development server
pnpm dev
Navigate to http://localhost:3000 and start commanding your APIs like you own the place.