feat(core): Rename vercelai.schema to gen_ai.request.schema

dev-packages/node-integration-tests/suites/tracing/vercelai/scenario-generate-object.mjs

@@ -0,0 +1,30 @@
+import * as Sentry from '@sentry/node';
+import { generateObject } from 'ai';
+import { MockLanguageModelV1 } from 'ai/test';
+import { z } from 'zod';
+
+async function run() {
+  await Sentry.startSpan({ op: 'function', name: 'main' }, async () => {
+    // Test generateObject with schema
+    await generateObject({
+      model: new MockLanguageModelV1({
+        defaultObjectGenerationMode: 'json',
+        doGenerate: async () => ({
+          rawCall: { rawPrompt: null, rawSettings: {} },
+          finishReason: 'stop',
+          usage: { promptTokens: 15, completionTokens: 25 },
+          text: '{ "name": "John Doe", "age": 30 }',
+        }),
+      }),
+      schema: z.object({
+        name: z.string().describe('The name of the person'),
+        age: z.number().describe('The age of the person'),
+      }),
+      schemaName: 'Person',
+      schemaDescription: 'A person with name and age',
+      prompt: 'Generate a person object',
+    });
+  });
+}
+
+run();

dev-packages/node-integration-tests/suites/tracing/vercelai/test-generate-object.ts

@@ -0,0 +1,67 @@
+import { afterAll, describe, expect } from 'vitest';
+import { cleanupChildProcesses, createEsmAndCjsTests } from '../../../utils/runner';
+
+describe('Vercel AI integration - generateObject', () => {
+  afterAll(() => {
+    cleanupChildProcesses();
+  });
+
+  const EXPECTED_TRANSACTION = {
+    transaction: 'main',
+    spans: expect.arrayContaining([
+      // generateObject span
+      expect.objectContaining({
+        data: expect.objectContaining({
+          'vercel.ai.model.id': 'mock-model-id',
+          'vercel.ai.model.provider': 'mock-provider',
+          'vercel.ai.operationId': 'ai.generateObject',
+          'vercel.ai.pipeline.name': 'generateObject',
+          'vercel.ai.streaming': false,
+          'vercel.ai.settings.mode': 'json',
+          'vercel.ai.settings.output': 'object',
+          'gen_ai.request.schema': expect.any(String),
+          'gen_ai.response.model': 'mock-model-id',
+          'gen_ai.usage.input_tokens': 15,
+          'gen_ai.usage.output_tokens': 25,
+          'gen_ai.usage.total_tokens': 40,
+          'operation.name': 'ai.generateObject',
+          'sentry.op': 'gen_ai.invoke_agent',
+          'sentry.origin': 'auto.vercelai.otel',
+        }),
+        description: 'generateObject',
+        op: 'gen_ai.invoke_agent',
+        origin: 'auto.vercelai.otel',
+        status: 'ok',
+      }),
+      // generateObject.doGenerate span
+      expect.objectContaining({
+        data: expect.objectContaining({
+          'sentry.origin': 'auto.vercelai.otel',
+          'sentry.op': 'gen_ai.generate_object',
+          'operation.name': 'ai.generateObject.doGenerate',
+          'vercel.ai.operationId': 'ai.generateObject.doGenerate',
+          'vercel.ai.model.provider': 'mock-provider',
+          'vercel.ai.model.id': 'mock-model-id',
+          'vercel.ai.pipeline.name': 'generateObject.doGenerate',
+          'vercel.ai.streaming': false,
+          'gen_ai.system': 'mock-provider',
+          'gen_ai.request.model': 'mock-model-id',
+          'gen_ai.response.model': 'mock-model-id',
+          'gen_ai.usage.input_tokens': 15,
+          'gen_ai.usage.output_tokens': 25,
+          'gen_ai.usage.total_tokens': 40,
+        }),
+        description: 'generate_object mock-model-id',
+        op: 'gen_ai.generate_object',
+        origin: 'auto.vercelai.otel',
+        status: 'ok',
+      }),
+    ]),
+  };
+
+  createEsmAndCjsTests(__dirname, 'scenario-generate-object.mjs', 'instrument.mjs', (createRunner, test) => {
+    test('captures generateObject spans with schema attributes', async () => {
+      await createRunner().expect({ transaction: EXPECTED_TRANSACTION }).start().completed();
+    });
+  });
+});

packages/core/src/utils/vercel-ai/index.ts

@@ -17,6 +17,7 @@ import {
   AI_RESPONSE_PROVIDER_METADATA_ATTRIBUTE,
   AI_RESPONSE_TEXT_ATTRIBUTE,
   AI_RESPONSE_TOOL_CALLS_ATTRIBUTE,
+  AI_SCHEMA_ATTRIBUTE,
   AI_TELEMETRY_FUNCTION_ID_ATTRIBUTE,
   AI_TOOL_CALL_ARGS_ATTRIBUTE,
   AI_TOOL_CALL_ID_ATTRIBUTE,
@@ -125,6 +126,8 @@ function processEndedVercelAiSpan(span: SpanJSON): void {
   renameAttributeKey(attributes, AI_TOOL_CALL_ARGS_ATTRIBUTE, 'gen_ai.tool.input');
   renameAttributeKey(attributes, AI_TOOL_CALL_RESULT_ATTRIBUTE, 'gen_ai.tool.output');
 
+  renameAttributeKey(attributes, AI_SCHEMA_ATTRIBUTE, 'gen_ai.request.schema');
+
   addProviderMetadataToAttributes(attributes);
 
   // Change attributes namespaced with `ai.X` to `vercel.ai.X`