AI Integration Expert | Bootspring Docs

The AI Integration Expert agent specializes in AI/LLM integration, Anthropic Claude SDK, AI SDK, streaming responses, and building AI-powered features.

Expertise Areas#

Anthropic Claude Integration - SDK setup, message handling
Streaming Responses - Real-time AI output streaming
Vercel AI SDK - Simplified AI integration patterns
Structured Output - JSON extraction and parsing
Tool Use / Function Calling - AI-driven tool execution
Rate Limiting - API usage management
Prompt Engineering - Effective prompt design
Cost Optimization - Token management and efficiency

Usage Examples#

Claude Integration#

Use the ai-integration-expert agent to set up Anthropic Claude integration for a Next.js chat application.

Response includes:

SDK initialization
Message handling
System prompts
Error handling

Streaming Chat#

Use the ai-integration-expert agent to implement a streaming chat interface with real-time responses.

Response includes:

API route with streaming
Client-side stream handling
UI component structure
Loading states

Tool Use#

Use the ai-integration-expert agent to implement function calling for a customer service bot.

Response includes:

Tool definitions
Execution handlers
Response continuation
Error handling

Best Practices Applied#

1. API Integration#

Secure API key handling
Proper error management
Response type checking
Rate limiting implementation

2. Streaming#

Efficient stream handling
Progress indicators
Graceful error recovery
Connection management

3. Structured Output#

Zod schema validation
JSON parsing safety
Type inference
Fallback handling

4. Cost Management#

Token counting
Usage tracking
Rate limiting
Caching strategies

Common Patterns#

Anthropic Claude Setup#

// lib/anthropic.ts
import Anthropic from '@anthropic-ai/sdk';

export const anthropic = new Anthropic({
  apiKey: process.env.ANTHROPIC_API_KEY,
});

export async function chat(prompt: string) {
  const response = await anthropic.messages.create({
    model: 'claude-sonnet-4-20250514',
    max_tokens: 1024,
    messages: [{ role: 'user', content: prompt }],
  });

  return response.content[0].type === 'text'
    ? response.content[0].text
    : null;
}

Streaming API Route#

// app/api/chat/route.ts
import { anthropic } from '@/lib/anthropic';

export async function POST(req: Request) {
  const { messages } = await req.json();

  const stream = await anthropic.messages.stream({
    model: 'claude-sonnet-4-20250514',
    max_tokens: 1024,
    messages,
  });

  return new Response(stream.toReadableStream(), {
    headers: {
      'Content-Type': 'text/event-stream',
      'Cache-Control': 'no-cache',
      Connection: 'keep-alive',
    },
  });
}

Vercel AI SDK Integration#

// app/api/chat/route.ts
import { anthropic } from '@ai-sdk/anthropic';
import { streamText } from 'ai';

export async function POST(req: Request) {
  const { messages } = await req.json();

  const result = streamText({
    model: anthropic('claude-sonnet-4-20250514'),
    messages,
  });

  return result.toDataStreamResponse();
}

Tool Use Implementation#

const tools = [
  {
    name: 'get_weather',
    description: 'Get the current weather for a location',
    input_schema: {
      type: 'object',
      properties: {
        location: {
          type: 'string',
          description: 'City and state, e.g., San Francisco, CA',
        },
      },
      required: ['location'],
    },
  },
];

export async function chatWithTools(prompt: string) {
  const response = await anthropic.messages.create({
    model: 'claude-sonnet-4-20250514',
    max_tokens: 1024,
    tools,
    messages: [{ role: 'user', content: prompt }],
  });

  // Handle tool use in response
  for (const block of response.content) {
    if (block.type === 'tool_use') {
      const result = await executeTool(block.name, block.input);
      // Continue conversation with tool result
    }
  }

  return response;
}

Sample Prompts#

Task	Prompt
Chat setup	"Set up a basic chat interface with Claude"
Streaming	"Implement streaming responses in a chat application"
Tool use	"Add function calling for database queries"
Rate limiting	"Implement rate limiting for AI API calls"
Structured output	"Extract structured data from user messages"

Configuration#

// bootspring.config.js
module.exports = {
  agents: {
    customInstructions: {
      'ai-integration-expert': `
        - Use Anthropic Claude as primary LLM
        - Implement streaming for better UX
        - Include rate limiting
        - Handle errors gracefully
        - Track token usage for cost management
      `,
    },
  },
  ai: {
    provider: 'anthropic',
    model: 'claude-sonnet-4-20250514',
    streaming: true,
  },
};

Backend Expert - API implementation
Frontend Expert - UI components
Performance Expert - Optimization