AgentForge: Autonomous Agent Builder & Red-Team Security Lab

An enterprise AI Agent workflow visual builder, execution engine, and automated adversarial red-teaming lab powered by Gemini and TypeScript.

---

🚀 Quick Start (Local Development)

1. Prerequisites

• Node.js: v18.0.0 or higher
• npm or pnpm
• Gemini API Key: Get a free API key from Google AI Studio

2. Clone & Install Dependencies

# Clone the repository
git clone https://github.com/YOUR_USERNAME/agentforge.git
cd agentforge

# Install packages
npm install

3. Configure Environment Variables

Create a .env file in the root directory:

cp .env.example .env

Inside .env, add your Gemini API Key:

GEMINI_API_KEY=your_gemini_api_key_here
PORT=3000

4. Run the Development Server

npm run dev

Open http://localhost:3000 in your browser.

---

🏗️ Production Build & Run

To build the optimized client bundle and server bundle:

# Build frontend and compile backend
npm run build

# Launch production server
npm start

The server will bind to 0.0.0.0:3000 (or the PORT environment variable) serving both the REST APIs and static SPA.

---

🐳 Deploying to Cloud & Hosting Platforms

Option A: Google Cloud Run (Recommended)

1. In Google Cloud Console or using gcloud:

gcloud run deploy agentforge \
  --source . \
  --platform managed \
  --region us-central1 \
  --allow-unauthenticated \
  --set-env-vars GEMINI_API_KEY="your_api_key_here"

Option B: Render / Railway / Fly.io

1. Connect your GitHub repository to Render or Railway.
2. Set the Build Command:

   npm install && npm run build

3. Set the Start Command:

   npm start

4. Add the Environment Variable GEMINI_API_KEY in your provider's dashboard.

---

📂 Project Architecture

• server.ts - Express backend proxying Gemini API calls, executing dynamic workflows, and providing /api/v1/agents/:id/run REST endpoints and /api/v1/security/evaluate-probe test runner.
• src/components/Canvas.tsx - Interactive visual node canvas for building multi-agent DAGs.
• src/components/SecurityAttackLab.tsx - Automated Red-Team adversarial attack lab mapped to OWASP LLM Top 10 & NIST AI RMF.
• src/components/DeployModal.tsx - Production deployment console with live cURL testing and webhook endpoints.