embr-foundry-chat-sample-node

A minimal Node.js / Express chat app that runs on Embr and uses an Azure AI Foundry-deployed model as its LLM. The agent loop is orchestrated locally via the Vercel AI SDK (ai + @ai-sdk/openai-compatible); only the LLM inference call reaches Foundry.

This is the Node sibling of embr-foundry-chat-sample-python and embr-foundry-chat-sample-dotnet. Same UI, same tools, same flows — different stack.

What's in the box

• ESM Node app on Node 20+ / Express 5.
• Two local tools the agent can call: getWeather and rollDice — defined with tool() from the AI SDK + Zod schemas.
• Foundry model is configured through the OpenAI-compatible endpoint via createOpenAICompatible({ baseURL, apiKey }), so the same code path works against Azure OpenAI or Foundry without any SDK swap.
• Multi-step tool-call loop courtesy of generateText({ stopWhen: stepCountIs(5) }).
• Single static public/index.html chat UI (markdown via marked + DOMPurify from CDN, no bundler).

Layout

app/
  server.js            # Express endpoints, message history, static mount
  agent.js             # Vercel AI SDK agent + Foundry client construction
  tools.js             # local tools the agent can invoke
  public/index.html    # chat UI
embr.yaml              # Embr deploy config (platform: nodejs)
.env.example           # required env vars (copy to .env for local runs)

Configuration

Three env vars (see .env.example):

Variable	Purpose
FOUNDRY_BASE_URL	OpenAI-compatible endpoint for your Foundry project, e.g. https://<resource>.services.ai.azure.com/api/projects/<project>/openai/v1
FOUNDRY_API_KEY	API key from the Foundry resource (Keys + Endpoint page)
FOUNDRY_MODEL_DEPLOYMENT	Deployment name, e.g. gpt-4o-mini, gpt-5.4-mini-1

Run locally

cp .env.example .env  # then fill in real values
npm install
npm start

Then open http://localhost:5000.

Smoke test:

curl http://localhost:5000/health
curl http://localhost:5000/api/config
curl -X POST http://localhost:5000/api/chat \
  -H 'Content-Type: application/json' \
  -d '{"message":"Roll a 20-sided die"}'

Deploy to Embr

gh repo create embr-foundry-chat-sample-node --source=. --public --push
embr quickstart deploy <your-user>/embr-foundry-chat-sample-node
embr variables set FOUNDRY_BASE_URL ... -p <id> -e <id> -s
embr variables set FOUNDRY_API_KEY ...  -p <id> -e <id> -s
embr variables set FOUNDRY_MODEL_DEPLOYMENT ... -p <id> -e <id> -s

The first deploy provisions the project + environment and triggers a build. Subsequent pushes auto-deploy. Note: setting variables on a live deployment requires embr deployments trigger to pick them up.

Endpoints

Path	Method	Purpose
/	GET	Chat UI (static public/index.html)
/health	GET	Liveness probe
/api/config	GET	Resolved Foundry endpoint summary (host/project/model)
/api/chat	POST {message, threadId?}	Run one agent turn
/api/reset	POST ?thread_id=…	Clear server-side history for a thread

Notes

• Agent and message history are in-memory — fine for the demo, not durable.
• Errors from the model surface as HTTP 502 with the provider message in detail. The user message is rolled back on failure so retries don't compound.
• Same conversation shape as the Python and .NET samples ({role, content}).