⚡ ApiforgeX
The Autonomous Backend Engineer.
Turn a single sentence into a production-ready, deployed API in 60 seconds.
View Demo Video | Report Bug
🚀 What is ApiforgeX?
ApiforgeX is not just a code generator; it's an Agentic DevOps Pipeline in a Box.
Backend development involves repetitive boilerplate: setting up Express, configuring Prisma, writing CRUD controllers, and managing deployments. ApiforgeX automates 100% of this.
✨ Key Features (The "Product" Promise)
🗣️ Natural Language Input: Just say "I need a blog API with posts and comments".
🧠 AI Architect (Oumi): Our specialized model designs a strict, valid Database Schema (no hallucinations).
🏗️ Auto-Scaffolding: Generates TypeScript, Express, Prisma, and Docker configurations instantly.
🤖 Automated Code Review: Every line is audited by CodeRabbit AI before merging.
🚀 Instant Deployment: Automatically deploys to Vercel serverless functions.
🔔 Real-time Updates: Get Slack notifications for every build step.
🛠️ How It Works (The Architecture)
ApiforgeX combines the power of best-in-class developer tools:
Interface: Cline (VS Code) acts as the pilot.
Orchestrator: Kestra manages the lifecycle (Generate -> Test -> Deploy).
Intelligence: Ollama (Llama 3.2) generates the logic.
Quality: CodeRabbit ensures code health.
Hosting: Vercel serves the API globally.
⚡ Quick Start
Prerequisites
Node.js 18+
Docker & Docker Compose (for Kestra)
GitHub Token & Vercel Token
Important: You must install the CodeRabbit GitHub App on your target repository for the automated review cycle to work.
The easiest way to use ApiforgeX is to install it globaly via npm:
npm install -g apiforgexOnce installed, you can generate a new backend project in seconds:
-
Initialize Configuration (Optional, for GitHub/Slack integration)
apiforgex init
-
Generate a Project
apiforgex generate
Follow the interactive prompts to describe your API, choose a database, and watch the magic happen!
ApiforgeX uses AI to generate code. You have two options:
- Google Gemini (Recommended): Set
GEMINI_API_KEYin your environment variables. - Local Ollama: Install Ollama and run
ollama run llama3.2.
If you want to contribute to the ApiforgeX source code, follow these steps:
-
Clone the repository
git clone https://github.com/shubham-01-star/ApiforgeX.git cd ApiforgeX npm install -
Start the Orchestrator
Launch Kestra in a local container:
docker-compose up -d
Visit http://localhost:8080 to see the dashboard.
- Run the Agent
You can run the CLI directly or trigger it via Kestra:
npm start -- generate --prompt "A task management app" --db postgresql
🏆 Hackathon Tracks
We are proudly building on top of:
Sponsor
Integration Details
Kestra Orchestrates the entire CI/CD pipeline, manages Docker containers, and handles Slack notifications. The heart of our autonomous workflow.
Cline acts as the pilot and primary interface. We built a custom MCP Server (
src/mcp/server.ts) to allow Cline to trigger complex Kestra workflows using natural language.
Oumi is used for Data Synthesis. We capture the logs and prompts from our agents to build a fine-tuning dataset, preparing for future self-improvement cycles.
CodeRabbit is our autonomous QA engineer. It reviews every Pull Request generated by the agent. Our "Watcher" script (
watcher.ts) parses CodeRabbit's feedback to auto-trigger fixes.
Vercel provides instant serverless deployment. Every successful merge in our pipeline automatically deploys the generated API to production.
🤝 Contributing
To enable the Agentic capabilities in VS Code:
- Open Cline Settings -> MCP Servers.
- Copy the content from
cline_mcp_config.jsonin this repo. - Paste it into your Cline MCP Settings file.
- Update the Environment Variables (GITHUB_TOKEN, SLACK_WEBHOOK etc.) with your real credentials.
- Ensure the project is built:
npm run build.
Now you can ask Cline:
"Generate a new library management project."