Tutorial: Integrating LLM Council APIs into Your Application

API Integration Overview

SPRAPP provides RESTful APIs for integrating LLM council capabilities into your applications. This tutorial covers everything you need.

Authentication

Getting Your API Key

1. Log into SPRAPP
2. Navigate to Settings > API
3. Generate a new API key
4. Store securely (never commit to git)

Using the Key

curl -X POST https://api.sprapp.com/v1/council/query \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "Explain quantum computing"}'

Basic Query

Request Format

{
  "query": "Your question here",
  "council": "default",
  "mode": "consensus",
  "models": ["claude-3.5-sonnet", "gpt-4o", "gemini-1.5-pro"]
}

Response Format

{
  "response": "The synthesized answer...",
  "consensus": true,
  "confidence": 0.94,
  "model_responses": [...],
  "latency_ms": 3420
}

Integration Examples

Python

import requests

def query_council(question):
    response = requests.post(
        "https://api.sprapp.com/v1/council/query",
        headers={"Authorization": f"Bearer {API_KEY}"},
        json={"query": question, "mode": "consensus"}
    )
    return response.json()

result = query_council("What is machine learning?")
print(result["response"])

JavaScript/Node.js

const response = await fetch('https://api.sprapp.com/v1/council/query', {
  method: 'POST',
  headers: {
    'Authorization': `Bearer ${API_KEY}`,
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({ query: 'Your question', mode: 'consensus' })
});
const data = await response.json();

Error Handling

Rate Limits

• Free tier: 100 requests/day
• Lite tier: 1000 requests/day
• Pro tier: Unlimited

Handle 429 errors with exponential backoff.

Timeouts

Set reasonable timeouts:

• Minimum: 30 seconds
• Recommended: 60 seconds
• Long queries: 120 seconds

Best Practices

1. Cache frequent queries
2. Implement retry logic
3. Log responses for debugging
4. Monitor API usage
5. Use streaming for long responses

Your application is now ready to leverage the power of LLM councils!