Reviews Summariser for CookinGenie.

Foodie Impressions is a full‑stack application that demonstrates an Angular frontend and a Node/Express backend used to collect chef reviews and produce short AI-generated summaries. The server proxies requests to an external Foundry/OpenAI-style API to generate summaries.

Key parts:

• Client: Angular app (client/angular.json, client/package.json)
• Server: Node + Express proxy and static server (server/index.js)
• CI/CD: GitHub Actions workflow for build & Azure deployment (.github/workflows/main_foodieimpressions.yml)

• Browse chefs and read/save reviews.
• Add new reviews (stored in localStorage).
• Automatic summary generation using the backend AI proxy.
• Static build served by the Node server for production.

• Node.js (recommended 18+)
• npm
• Angular CLI (for development): npm i -g @angular/cli

The server expects the following environment variables (see server/index.js):

• FOUNDRY_BASE — base URL for Foundry/OpenAI proxy target
• FOUNDRY_API_KEY — API key for the Foundry/OpenAI service

Set these in a .env file in the server/ folder or provide them in your production environment. The GitHub Actions workflow uses Azure secrets for deployment credentials.

1. Install dependencies

   • Client: cd client && npm ci
   • Server: cd server && npm ci

2. Run client (dev)

   • cd client
   • npm start
   • Opens on http://localhost:4200 and uses proxy config for API calls.

3. Run server (dev)

   • cd server
   • node index.js
   • Server serves static files from the built client (see public path in server/index.js) and exposes API endpoints such as POST /api/ai/responses.

4. To use the full app locally:

   • Build the client so assets are available to the server:
     • cd client && npm run build
     • This writes build artifacts into the server/public folder per client/angular.json.
   • Start the server:
     • cd server && node index.js
   • Navigate to http://localhost:3000 (or the port configured in server).

• Client build: cd client && npm run build
  • Output path is configured in client/angular.json to place files under ../server/public.
• CI/CD: The repo contains a GitHub Actions workflow that builds client and server and deploys to Azure Web App (.github/workflows/main_foodieimpressions.yml).

• POST /api/ai/responses
  • Forwards structured requests to the Foundry/OpenAI responses endpoint. See server/index.js for request structure and proxy behavior.
  • Example (curl): curl -X POST http://localhost:3000/api/ai/responses -H "Content-Type: application/json" -d '{"input":[{"role":"user","content":"..."}],"model":"gpt-5-mini"}'

• Frontend summarization logic: client/src/app/components/chef-profile/chef-profile.component.ts
• OpenAI client service (frontend): client/src/app/services/openai.service.ts
• Server entrypoint: server/index.js
• Angular build config: client/angular.json

• Open issues and PRs.
• Keep changes small and focused; update README and any environment or deployment docs if you modify server behavior or build outputs.