ai-sdk/azure does not call tools

[patolax]

I modified the Angular example to interface with Azure OpenAI model. However, I can only get general conversation responses, if I try to force it to do a tool call (e.g., "Get the weather in DC" ), it returns empty. Am I doing something wrong here?

`import { createAzure } from '@ai-sdk/azure';
import { convertToModelMessages, streamObject, streamText, tool } from 'ai';
import 'dotenv/config';
import express, { Request, Response } from 'express';
import z from 'zod';
import { experimental_createMCPClient as createMCPClient } from 'ai';
import { Experimental_StdioMCPTransport as StdioMCPTransport } from 'ai/mcp-stdio';
import { StreamableHTTPClientTransport } from '@modelcontextprotocol/sdk/client/streamableHttp';

async function startServer() {

const azure = createAzure({
resourceName: 'ai-sdk-test', // extracted from your endpoint
apiKey: process.env['AZURE_OPENAI_API_KEY'],
});

const app = express();
app.use(express.json({ strict: false })); // Allow primitives (for analyze endpoint)

app.post('/api/chat', async (req: Request, res: Response) => {
const { messages, selectedModel } = req.body;
const result = streamText({
model: azure(selectedModel || 'gpt-4.1'), // Use your deployment name
messages: convertToModelMessages(messages),
tools: allTools,
});

result.pipeUIMessageStreamToResponse(res);

});

app.post('/api/completion', async (req: Request, res: Response) => {
try {
const { prompt } = req.body;
console.log('Received completion request:', { prompt });

  const result = streamText({
    model: azure('gpt-4.1'),
    tools: {
      weather: tool({
        description: 'Get the weather in a location',
        inputSchema: z.object({
          location: z.string().describe('The location to get the weather for'),
        }),
        execute: async ({ location }) => ({
          location,
          temperature: 72 + Math.floor(Math.random() * 21) - 10,
        }),
      }),

    },
    prompt,

  });

  console.log('Streaming response...');
  result.pipeTextStreamToResponse(res);
} catch (error) {
  console.error('Error in completion endpoint:', error);
  res.status(500).json({ error: 'Internal server error' });
}

});

app.post('/api/analyze', express.raw(), async (req: Request, res: Response) => {
const input = req.body.toString('utf8');

const result = streamObject({
  model: azure('gpt-4.1'), // Use your deployment name
  schema: z.object({
    title: z.string(),
    summary: z.string(),
    tags: z.array(z.string()),
    sentiment: z.enum(['positive', 'negative', 'neutral']),
  }),
  prompt: `Analyze this content: ${input}`,
});

result.pipeTextStreamToResponse(res);

});

app.listen(3000, () => {
console.log(Example app listening on port ${3000});
});
}

// Start the server
startServer().catch(console.error);
`