NameForge

AI-powered project name generator. Describe your project, pick a style, and get 6 memorable name suggestions — scored and explained.

Built with React + TypeScript (frontend) and Node.js + Express (backend). Supports multiple AI providers.

---

AI Providers

LLM_PROVIDER	Description	Requires
claude-code	Runs via Claude Code session (default)	Claude Code installed locally
anthropic-api	Calls Anthropic API directly	ANTHROPIC_API_KEY
azure-foundry	Azure AI Foundry / Azure OpenAI	AZURE_FOUNDRY_ENDPOINT, AZURE_FOUNDRY_API_KEY
ollama	Local LLM via Ollama	Ollama running locally

---

Local Development

Prerequisites

• Node.js 20+
• One of the AI providers above

Setup

git clone https://github.com/LeePepe/NameForge.git
cd NameForge

# Install all dependencies
npm install

# Configure environment
cp backend/.env.example backend/.env
# Edit backend/.env and set LLM_PROVIDER + any required keys

Run

# Start both frontend and backend with hot-reload
npm run dev

• Frontend: http://localhost:5173
• Backend: http://localhost:3001

---

Docker (Self-hosted)

Anthropic API mode

ANTHROPIC_API_KEY=sk-ant-xxx docker compose --profile anthropic up

Open http://localhost:3001

Ollama mode (fully local, no API key)

docker compose --profile ollama up

# Pull a model (first time only)
docker exec -it nameforge-ollama-1 ollama pull llama3.2

Open http://localhost:3001

Azure Foundry mode

AZURE_FOUNDRY_ENDPOINT=https://your-resource.cognitiveservices.azure.com/ \
AZURE_FOUNDRY_API_KEY=your-key \
docker compose --profile azure-foundry up

Open http://localhost:3001

---

Deploy to Azure Container Apps

Production deploys now target Azure Container Apps in Southeast Asia (Singapore) via GitHub Actions.

1. Create these GitHub repository secrets:
   • AZURE_CREDENTIALS
   • AZURE_FOUNDRY_API_KEY
2. Create these GitHub repository variables:
   • AZURE_FOUNDRY_ENDPOINT
   • AZURE_FOUNDRY_DEPLOYMENT (optional, defaults to Kimi-K2.5)
   • AZURE_FOUNDRY_API_VERSION (optional, defaults to 2024-12-01-preview)
3. Push to main
4. The workflow will:
   • provision or update the Singapore Log Analytics workspace, Container Apps environment, ACR, and Container App from infra/main.bicep
   • build the Docker image in the Singapore ACR
   • deploy the new image to the Singapore Container App

The deployment keeps the existing NameForge resource group, but creates a separate Singapore application stack so it can coexist with the older East US deployment during migration.

---

Deploy to Railway / Render

These platforms support Docker deployments out of the box.

Railway:

railway up

Render: Create a new Web Service, point to this repo, select "Docker" as runtime.

Set these environment variables on the platform:

LLM_PROVIDER=anthropic-api
ANTHROPIC_API_KEY=sk-ant-xxx
SERVE_STATIC=true

---

Environment Variables

Variable	Default	Description
LLM_PROVIDER	claude-code	AI provider: claude-code, anthropic-api, azure-foundry, ollama
ANTHROPIC_API_KEY	—	Required when LLM_PROVIDER=anthropic-api
ANTHROPIC_MODEL	claude-opus-4-5	Anthropic model to use
AZURE_FOUNDRY_ENDPOINT	—	Required when LLM_PROVIDER=azure-foundry (e.g. https://your-resource.cognitiveservices.azure.com/)
AZURE_FOUNDRY_API_KEY	—	Required when LLM_PROVIDER=azure-foundry
AZURE_FOUNDRY_DEPLOYMENT	Kimi-K2.5	Azure Foundry model deployment name
AZURE_FOUNDRY_API_VERSION	2024-12-01-preview	Azure OpenAI API version
OLLAMA_BASE_URL	http://localhost:11434	Ollama server URL
OLLAMA_MODEL	llama3.2	Ollama model to use
PORT	3001	Backend server port
FRONTEND_URL	http://localhost:5173	Allowed CORS origin (dev mode)
SERVE_STATIC	false	Serve frontend from Express (Docker mode)

---

Project Structure

NameForge/
├── frontend/          # React + TypeScript + Vite
│   └── src/
│       ├── App.tsx
│       ├── api/       # API client
│       └── components/
├── backend/           # Node.js + Express + TypeScript
│   └── src/
│       ├── index.ts
│       ├── routes/
│       └── services/
│           └── claude.ts   # AI provider abstraction
├── Dockerfile         # Single-container production build
└── docker-compose.yml # Multi-profile orchestration

---

License

MIT — see LICENSE