AI Functions Examples

The examples/ai-functions/ directory contains scripts for validating, testing, and iterating on AI SDK functions across providers.

Example Categories

Examples are organized by AI SDK function in examples/ai-functions/src/:

File Naming Convention

Examples follow the pattern: {provider}-{feature}.ts

Example Structure

All examples use the run() wrapper from lib/run.ts which:

• Loads environment variables from .env
• Provides error handling with detailed API error logging

Basic Template

typescript
1import { providerName } from '@ai-sdk/provider-name';
2import { generateText } from 'ai';
3import { run } from '../lib/run';
4
5run(async () => {
6  const result = await generateText({
7    model: providerName('model-id'),
8    prompt: 'Your prompt here.',
9  });
10
11  console.log(result.text);
12  console.log('Token usage:', result.usage);
13  console.log('Finish reason:', result.finishReason);
14});

Streaming Template

typescript
1import { providerName } from '@ai-sdk/provider-name';
2import { streamText } from 'ai';
3import { printFullStream } from '../lib/print-full-stream';
4import { run } from '../lib/run';
5
6run(async () => {
7  const result = streamText({
8    model: providerName('model-id'),
9    prompt: 'Your prompt here.',
10  });
11
12  await printFullStream({ result });
13});

Tool Calling Template

typescript
1import { providerName } from '@ai-sdk/provider-name';
2import { generateText, tool } from 'ai';
3import { z } from 'zod';
4import { run } from '../lib/run';
5
6run(async () => {
7  const result = await generateText({
8    model: providerName('model-id'),
9    tools: {
10      myTool: tool({
11        description: 'Tool description',
12        inputSchema: z.object({
13          param: z.string().describe('Parameter description'),
14        }),
15        execute: async ({ param }) => {
16          return { result: `Processed: ${param}` };
17        },
18      }),
19    },
20    prompt: 'Use the tool to...',
21  });
22
23  console.log(JSON.stringify(result, null, 2));
24});

Structured Output Template

typescript
1import { providerName } from '@ai-sdk/provider-name';
2import { generateObject } from 'ai';
3import { z } from 'zod';
4import { run } from '../lib/run';
5
6run(async () => {
7  const result = await generateObject({
8    model: providerName('model-id'),
9    schema: z.object({
10      name: z.string(),
11      items: z.array(z.string()),
12    }),
13    prompt: 'Generate a...',
14  });
15
16  console.log(JSON.stringify(result.object, null, 2));
17  console.log('Token usage:', result.usage);
18});

Running Examples

From the examples/ai-functions directory:

bash
1pnpm tsx src/generate-text/openai.ts
2pnpm tsx src/stream-text/openai-tool-call.ts
3pnpm tsx src/agent/openai-generate.ts

When to Write Examples

Write examples when:

1. Adding a new provider: Create basic examples for each supported API (generateText, streamText, generateObject, etc.)

2. Implementing a new feature: Demonstrate the feature with at least one provider example

3. Reproducing a bug: Create an example that shows the issue for debugging

4. Adding provider-specific options: Show how to use providerOptions for provider-specific settings

5. Creating test fixtures: Use examples to generate API response fixtures (see capture-api-response-test-fixture skill)

Utility Helpers

The lib/ directory contains shared utilities:

Using print utilities

typescript
1import { print } from '../lib/print';
2
3// Pretty print objects without undefined values
4print('Result:', result);
5print('Usage:', result.usage, { depth: 2 });

Using printFullStream

typescript
1import { printFullStream } from '../lib/print-full-stream';
2
3const result = streamText({ ... });
4await printFullStream({ result }); // Colored output for text, tool calls, reasoning

Reusable Tools

The tools/ directory contains reusable tool definitions:

typescript
1import { weatherTool } from '../tools/weather-tool';
2
3const result = await generateText({
4  model: openai('gpt-4o'),
5  tools: { weather: weatherTool },
6  prompt: 'What is the weather in San Francisco?',
7});

Best Practices

1. Keep examples focused: Each example should demonstrate one feature or use case

2. Use descriptive prompts: Make it clear what the example is testing

3. Handle errors gracefully: The run() wrapper handles this automatically

4. Use realistic model IDs: Use actual model IDs that work with the provider

5. Add comments for complex logic: Explain non-obvious code patterns

6. Reuse tools when appropriate: Use weatherTool or create new reusable tools in tools/