@relayplane/sdk v0.2

🪄 Zero-config AI for developers - The easiest way to add intelligent AI to your applications

RelayPlane SDK provides seamless AI integration with automatic optimization, intelligent model selection, and built-in examples. No configuration required - just ask!

📚 Complete Documentation →

✨ Zero-Config Wins (New in v0.2)

🪄 Zero-Config Quick Start

1import RelayPlane from '@relayplane/sdk';
2
3// Magic! Just ask - auto-detects your API keys and picks the best model
4const result = await RelayPlane.ask("Explain quantum computing simply");
5console.log(result.response);
6console.log(result.reasoning.rationale); // See why this model was chosen

What makes this magical:

• Auto-detects your available API keys
• Intelligently selects the best model for your task
• Provides reasoning for every decision
• Falls back gracefully if providers are unavailable
• Works instantly without any setup

🎯 Smart Model Selection with Reasoning

1// Get intelligent model selection with full reasoning
2const result = await RelayPlane.ask("Complex financial analysis task", {
3  budget: 'moderate',        // 'minimal', 'moderate', 'unlimited'
4  priority: 'quality',       // 'speed', 'balanced', 'quality'
5  taskType: 'analysis'       // 'coding', 'creative', 'analysis', 'general'
6});
7
8console.log(result.reasoning.rationale);
9// "Selected Claude-3-sonnet over GPT-4 because:
10//  - 40% faster for financial analysis tasks
11//  - $0.02 vs $0.08 estimated cost  
12//  - 94% success rate for this task type"

📚 Built-in Examples Library

Ready-to-use patterns for instant productivity:

1// Summarization with intelligent length control
2const summary = await RelayPlane.examples.summarize(longText, { 
3  length: 'brief' 
4});
5
6// Translation with context awareness  
7const translation = await RelayPlane.examples.translate('Hello world', 'Spanish');
8
9// Code review with focus areas
10const review = await RelayPlane.examples.codeReview(myCode, { 
11  focus: 'security' 
12});
13
14// Research workflow
15const research = await RelayPlane.examples.research('quantum computing applications');
16
17// Chatbot with personality
18const chat = await RelayPlane.examples.chatbot('How are you?', {
19  personality: 'helpful and friendly assistant',
20  context: previousMessages
21});

⚡ Smart Retry with Automatic Model Switching

1// Automatic retry with intelligent model switching
2const result = await RelayPlane.smartRetry({
3  to: 'claude-sonnet-4-20250514',
4  payload: { messages: [{ role: 'user', content: 'Complex task' }] }
5}, {
6  maxRetries: 3,
7  confidenceThreshold: 0.9, // Retry until 90% confident
8  retryOnFailureTypes: ['rate_limit', 'timeout', 'model_error']
9});
10
11console.log(result.retryInfo.modelAttempts); // See all attempts and reasoning

🌊 Optimized Streaming

1// Enhanced streaming with intelligent chunking
2for await (const { chunk, metadata } of RelayPlane.optimizeStreaming(response, 'claude-sonnet-4-20250514')) {
3  console.log(chunk); // Optimally sized chunks for natural reading
4  console.log(metadata.isTyping); // Typing indicators
5  console.log(metadata.estimatedCost); // Real-time cost tracking
6  console.log(metadata.estimatedProgress); // Progress estimation
7}
8
9// React integration (if using React)
10const { content, metadata, isStreaming, startStreaming } = RelayPlane.useOptimizedStreaming();

🔧 Contextual Error Recovery

1try {
2  const result = await RelayPlane.ask("Test prompt");
3} catch (error) {
4  if (error instanceof RelayPlane.RelayApiKeyError) {
5    console.log(error.getHelpfulMessage());
6    // Shows exact steps to fix API key issues with copy-paste commands
7  }
8  
9  if (error instanceof RelayPlane.RelayRateLimitError) {
10    console.log(error.suggestions);
11    // ["Use RelayPlane.configure({ optimization: true }) for automatic fallbacks",
12    //  "Switch to hosted mode for better rate limiting",
13    //  "Try again in 60 seconds"]
14  }
15}

🚀 Instant Demo Experience

Want to see all these features in action? Try our interactive CLI:

1# Experience the magic in 30 seconds
2npx @relayplane/cli demo

This provides a beautiful terminal experience showcasing:

• Zero-Config Wins
• Smart model selection reasoning
• Built-in examples in action
• Performance comparisons
• Error recovery demonstrations

📦 Installation

1npm install @relayplane/sdk

⚙️ Configuration (Optional!)

RelayPlane works without configuration, but you can customize it:

1// Optional: Configure for enhanced features
2RelayPlane.configure({
3  apiKey: 'your-relayplane-key',     // For hosted mode
4  debug: true,                       // Enable debug logging
5  defaultOptimization: {
6    enabled: true,                   // Auto-fallback and optimization
7    strategy: 'balanced'             // 'speed', 'balanced', 'quality'
8  },
9  costMonitoring: {
10    dailyLimit: 10.00,              // Daily spend limit
11    alertThreshold: 0.8             // Alert at 80% of limit
12  }
13});
14
15// Or use environment variables (zero config!)
16// RELAYPLANE_API_KEY=rp_... (hosted mode)
17// OPENAI_API_KEY=sk-...     (local mode)
18// ANTHROPIC_API_KEY=sk-...  (local mode)
19// GOOGLE_API_KEY=...        (local mode)

🔗 Advanced Usage

Multi-Agent Chaining

1// Create intelligent workflows
2const result = await RelayPlane.chain({
3  steps: [
4    { step: 'research', to: 'claude-sonnet-4-20250514', prompt: 'Research topic' },
5    { step: 'analyze', to: 'gpt-4-turbo', prompt: 'Analyze findings' },
6    { step: 'summarize', to: 'claude-haiku', prompt: 'Create summary' }
7  ],
8  input: 'Analyze the impact of AI on healthcare',
9  optimization: true
10});

Batch Processing

1// Process multiple requests efficiently
2const results = await RelayPlane.batch({
3  requests: [
4    { to: 'claude-sonnet-4-20250514', payload: { messages: [...] } },
5    { to: 'gpt-4-turbo', payload: { messages: [...] } }
6  ],
7  parallel: true,
8  optimization: { enabled: true }
9});

Manual Relay (Advanced Users)

1// Full control when needed
2const response = await RelayPlane.relay({
3  to: 'claude-sonnet-4-20250514',
4  payload: {
5    messages: [
6      { role: 'user', content: 'Hello, Claude!' }
7    ],
8    max_tokens: 1000,
9    temperature: 0.7
10  },
11  optimization: {
12    enabled: true,
13    fallback: ['gpt-4-turbo', 'claude-haiku']
14  }
15});

🎯 Why RelayPlane v0.2?

Before RelayPlane

1// Complex setup required
2import OpenAI from 'openai';
3import Anthropic from '@anthropic-ai/sdk';
4
5const openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });
6const anthropic = new Anthropic({ apiKey: process.env.ANTHROPIC_API_KEY });
7
8// Manual model selection and error handling
9let response;
10try {
11  response = await openai.chat.completions.create({
12    model: 'gpt-4',  // Hard-coded model choice
13    messages: [{ role: 'user', content: 'Hello' }]
14  });
15} catch (error) {
16  if (error.status === 429) {
17    // Manual fallback logic
18    response = await anthropic.messages.create({
19      model: 'claude-3-sonnet-20240229',
20      messages: [{ role: 'user', content: 'Hello' }]
21    });
22  }
23}

With RelayPlane v0.2

1// Zero config, intelligent optimization
2import RelayPlane from '@relayplane/sdk';
3
4// That's it! Everything else is automatic
5const result = await RelayPlane.ask('Hello');
6console.log(result.response.body);
7console.log(result.reasoning.rationale); // See the intelligent decision-making

Key Differentiators

✅ Zero Configuration - Works instantly without setup
✅ Intelligent Model Selection - AI chooses the best model for your task
✅ Built-in Examples - Ready-to-use patterns for common tasks
✅ Smart Retry Logic - Automatic fallback with reasoning
✅ Contextual Error Messages - Helpful recovery suggestions
✅ Real-time Cost Tracking - Know your spend as you go
✅ Optimized Streaming - Perfect chunk sizes and timing
✅ Dual Mode Operation - Use your keys or our hosted service

📚 Supported Models

Provider	Model Identifiers
Anthropic	claude-opus-4-20250514, claude-sonnet-4-20250514, claude-3-7-sonnet-20250219, claude-3-5-sonnet-20241022, claude-3-5-haiku-20241022, claude-3-opus-20240229, claude-3-sonnet-20240229, claude-3-haiku-20240307
OpenAI	gpt-4.1, gpt-4.1-mini, o3, o3-pro, o3-mini, o4-mini, gpt-4o, gpt-4o-mini, o1-preview, o1-mini, gpt-4, gpt-4-turbo, gpt-3.5-turbo
Google	gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, gemini-2.0-flash, gemini-1.5-pro, gemini-1.5-flash, gemini-pro, gemini-pro-vision

🔗 Links

• 📖 Full Documentation
• 🎮 Interactive Playground
• 🧪 Try the CLI Demo
• 💬 GitHub Discussions
• 🐙 GitHub Repository

🆘 Support

Quick Help

1# Test your setup instantly
2npx @relayplane/cli test
3
4# See error recovery demos
5npx @relayplane/cli demo
6# Select: "🔧 Error Recovery"

Get Help

• Documentation
• Troubleshooting Guide
• GitHub Discussions
• GitHub Issues

---

Start with zero config: npm install @relayplane/sdk 🪄