Reference

Complete reference documentation for NeuroLink configuration, troubleshooting, and technical details.

🎯 Reference Hub

This section provides comprehensive reference materials for advanced usage, configuration, and problem-solving.

Troubleshooting — Common issues, error messages, and solutions for NeuroLink CLI and SDK usage.
Configuration — Complete configuration reference including environment variables, provider settings, and optimization.
Provider Capabilities Audit — Capability matrix for the 13 text/multimodal providers it historically tracks — the other 27 are covered in the per-provider guides with capability matrices and configuration examples.
Provider Comparison — Detailed comparison of the hand-written provider implementations with features, costs, and recommendations.
FAQ — Frequently asked questions about NeuroLink features, limitations, and best practices.
Error Codes — Complete error code reference with categorized codes, severity levels, and resolution guidance.
Analytics — Comprehensive guide to NeuroLink analytics, metrics, token tracking, cost monitoring, and observability integration.
Telemetry Guide — OTLP setup, exporter behavior, and the local OpenObserve workflow for the Claude proxy.
Server Configuration — Configuration reference for server adapters including Hono, Express, Fastify, and Koa framework integration.
MCP Enhancements API — API reference for MCP enhancements including ToolRouter, ToolCache, RequestBatcher, tool annotations, and elicitation protocol.

🔧 Quick Reference

Environment Variables

# Core Provider API Keys
OPENAI_API_KEY="sk-your-openai-key"
GOOGLE_AI_API_KEY="AIza-your-google-ai-key"
ANTHROPIC_API_KEY="sk-ant-your-key"

# AWS Bedrock (requires AWS credentials)
AWS_ACCESS_KEY_ID="your-access-key"
AWS_SECRET_ACCESS_KEY="your-secret-key"
AWS_REGION="us-east-1"

# Azure OpenAI
AZURE_OPENAI_API_KEY="your-azure-key"
AZURE_OPENAI_ENDPOINT="https://your-resource.openai.azure.com"

# Google Vertex AI
GOOGLE_APPLICATION_CREDENTIALS="/path/to/service-account.json"

# Hugging Face
HUGGINGFACE_API_KEY="hf_your-key"

# Mistral AI
MISTRAL_API_KEY="your-mistral-key"

CLI Quick Commands

# Status and diagnostics
neurolink status                    # Check all providers
neurolink status --verbose         # Detailed diagnostics
neurolink provider status          # Provider-specific status

# Text generation
neurolink generate "prompt"        # Basic generation
neurolink gen "prompt" -p openai   # Specific provider
neurolink stream "prompt"          # Real-time streaming

# Configuration
neurolink config show              # Show current config
neurolink config validate          # Validate setup
neurolink config init              # Interactive setup

# MCP tools
neurolink mcp discover             # Find available servers
neurolink mcp list                 # List installed servers
neurolink mcp install <server>     # Install MCP server

# Claude proxy + local telemetry
neurolink proxy setup
neurolink proxy status --format json
neurolink proxy telemetry setup
neurolink proxy telemetry status

SDK Quick Reference

import { NeuroLink, createBestAIProvider } from "@juspay/neurolink";

// Basic usage
const neurolink = new NeuroLink();
const result = await neurolink.generate({
  input: { text: "Your prompt" },
  provider: "auto", // or specific provider
});

// Auto-select best provider
const provider = createBestAIProvider();
const result = await provider.generate({
  input: { text: "Your prompt" },
});

// With advanced options
const result = await neurolink.generate({
  input: { text: "Your prompt" },
  provider: "google-ai",
  model: "gemini-2.5-pro",
  temperature: 0.7,
  maxTokens: 1000,
  enableAnalytics: true,
  enableEvaluation: true,
  timeout: "30s",
});

📊 Provider Comparison Matrix

Quick Overview (see Provider Capabilities Audit for complete details):

Feature	OpenAI	Google AI	Anthropic	Bedrock	Azure	Vertex	HuggingFace	Ollama	Mistral	LiteLLM	SageMaker	OpenRouter	OpenAI Compat
Free Tier	❌	✅	❌	❌	❌	❌	✅	✅	✅	Varies	❌	Varies	Varies
Tool Support	✅	✅	✅	✅	✅	✅	⚠️	⚠️	✅	✅	✅	✅	✅
Streaming	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅
Vision	✅	✅	✅	✅	✅	✅	✅	⚠️	❌	✅	Varies	✅	Varies
Local	❌	❌	❌	❌	❌	❌	❌	✅	❌	❌	❌	❌	Varies
Enterprise	✅	✅	✅	✅	✅	✅	⚠️	✅	✅	✅	✅	✅	Varies

For detailed capability matrices, authentication requirements, and configuration examples, see:

Provider Capabilities Audit - Technical implementation details
Provider Comparison - Feature comparison and selection guide

🔍 Error Code Reference

Common Error Codes

Code	Description	Solution
`AUTH_ERROR`	Invalid API key or credentials	Check environment variables
`RATE_LIMIT`	API rate limit exceeded	Implement delays or upgrade plan
`TIMEOUT`	Request timeout	Increase timeout or check network
`MODEL_NOT_FOUND`	Invalid model name	Check available models
`TOOL_ERROR`	MCP tool execution failed	Check tool configuration
`PROVIDER_UNAVAILABLE`	Provider service down	Try different provider

Debugging Tips

# Enable debug mode
neurolink generate "test" --debug

# Verbose logging
neurolink status --verbose

# Check configuration
neurolink config validate

// SDK debugging
const neurolink = new NeuroLink({
  debug: true,
  logLevel: "verbose",
});

📈 Performance Optimization

Response Time Optimization

Provider selection: Use fastest providers for your region
Model selection: Choose appropriate model size for task
Concurrency: Limit parallel requests to avoid rate limits
Caching: Implement response caching for repeated queries

Cost Optimization

Model selection: Use cost-effective models when possible
Token management: Optimize prompt length and max tokens
Provider comparison: Compare costs across providers
Monitoring: Track usage with analytics

Memory Management

Streaming: Use streaming for large responses
Batch processing: Process multiple requests efficiently
Cleanup: Proper resource cleanup in long-running applications

🔐 Security Best Practices

API Key Management

Environment variables: Store keys in .env files
Never commit: Keep keys out of version control
Rotation: Regularly rotate API keys
Scope limitation: Use least-privilege access

Production Deployment

Secret management: Use secure secret management systems
Network security: Implement proper network controls
Monitoring: Log and monitor API usage
Error handling: Don't expose sensitive errors

🆘 Getting Help

Support Channels

GitHub Issues - Bug reports and feature requests
GitHub Discussions - Community questions
Documentation - Comprehensive guides and references
Examples - Practical implementation patterns

Before Asking for Help

Check the Troubleshooting Guide
Review the FAQ
Search existing GitHub Issues
Try the --debug flag for more information

Reporting Issues

When reporting issues, include:

NeuroLink version: npm list @juspay/neurolink
Node.js version: node --version
Operating system: OS and version
Error message: Complete error output
Reproduction steps: Minimal example to reproduce
Configuration: Relevant environment variables (without keys)

🔗 External Resources

AI Provider Documentation

OpenAI API - OpenAI official documentation
Google AI Studio - Google AI platform docs
Anthropic Claude - Anthropic API reference
AWS Bedrock - Amazon Bedrock guide

Vercel AI SDK - Separate framework NeuroLink interoperates with via the client SDK's LanguageModelV1 adapter
Model Context Protocol - Tool integration standard
TypeScript - Type safety and development

🎯 Reference Hub​

🔧 Quick Reference​

Environment Variables​

CLI Quick Commands​

SDK Quick Reference​

📊 Provider Comparison Matrix​

🔍 Error Code Reference​

Common Error Codes​

Debugging Tips​

📈 Performance Optimization​

Response Time Optimization​

Cost Optimization​

Memory Management​

🔐 Security Best Practices​

API Key Management​

Production Deployment​

🆘 Getting Help​

Support Channels​

Before Asking for Help​

Reporting Issues​

🔗 External Resources​

AI Provider Documentation​

Related Projects​