NeuroLink Usage Guide
NeuroLink is an enterprise AI development platform providing unified access to 30+ AI providers (text, voice, multimodal) through a single API. It ships as both a TypeScript SDK (@juspay/neurolink) and a professional CLI.
Quick Navigation
Based on your query, I'll guide you to the right documentation:
- Getting Started → Read sdk-quickstart.md
- Provider Setup → Read providers.md
- Multimodal (images, PDFs, files) → Read multimodal.md
- MCP Tools Integration → Read tools-mcp.md
- RAG Pipelines → Read rag-integration.md
- Conversation Memory → Read memory-conversations.md
- CLI Commands → Read cli-reference.md
- Advanced Features → Read advanced-features.md
- Troubleshooting → Read troubleshooting.md
Topic Routing
If the user asked about $ARGUMENTS:
| Topic Keywords |
Reference File |
| install, setup, start, begin, quickstart |
sdk-quickstart.md |
| provider, openai, anthropic, vertex, bedrock, azure, gemini, claude, model |
providers.md |
| image, pdf, csv, excel, document, file, multimodal, vision |
multimodal.md |
| tool, mcp, server, GitHub, external, function |
tools-mcp.md |
| rag, retrieval, chunk, vector, embed, document search |
rag-integration.md |
| memory, conversation, history, session, redis, context |
memory-conversations.md |
| cli, command, terminal, generate, stream, loop, serve |
cli-reference.md |
| hitl, workflow, agent, observe, telemetry, deploy, server |
advanced-features.md |
| error, issue, problem, fix, debug, not working |
troubleshooting.md |
Installation
npm install @juspay/neurolink
# or
pnpm add @juspay/neurolink
# or
yarn add @juspay/neurolink
Minimal Example
import { NeuroLink } from "@juspay/neurolink";
const neurolink = new NeuroLink();
// Generate text
const result = await neurolink.generate("Explain quantum computing");
console.log(result.content);
// Stream response
const stream = await neurolink.stream({ input: { text: "Write a story" } });
for await (const chunk of stream.stream) {
if ("content" in chunk) {
process.stdout.write(chunk.content);
}
}
Key Capabilities
| Feature |
Description |
| 30+ Providers |
OpenAI, Anthropic, Vertex, Bedrock, Azure, Mistral, Ollama, etc. |
| Multimodal |
Images, PDFs, CSV, Excel, Word, 50+ file types |
| MCP Tools |
58+ tools via Model Context Protocol |
| RAG |
Built-in chunking, embedding, vector search |
| Memory |
Conversation history with Redis support |
| Streaming |
Real-time token streaming |
| HITL |
Human-in-the-loop approval workflows |
| Observability |
Langfuse, OpenTelemetry integration |
Code Templates
Ready-to-use examples in templates/:
templates/basic-setup.ts - Basic SDK initializationtemplates/streaming.ts - Streaming responsestemplates/with-tools.ts - Tool integrationtemplates/rag-pipeline.ts - RAG usagetemplates/server-deploy.ts - HTTP server deployment
CLI Quick Reference
# Generate content
neurolink generate "Your prompt"
# Stream output
neurolink stream "Write a story"
# Interactive mode
neurolink loop
# Start HTTP server
neurolink serve --port 3000
# Setup providers
neurolink setup openai
Environment Variables
Set up your provider credentials:
# OpenAI
OPENAI_API_KEY=sk-...
# Anthropic
ANTHROPIC_API_KEY=sk-ant-...
# Google AI Studio
GOOGLE_API_KEY=...
# Vertex AI
VERTEX_PROJECT_ID=...
GOOGLE_APPLICATION_CREDENTIALS=/path/to/credentials.json
# AWS Bedrock
AWS_ACCESS_KEY_ID=...
AWS_SECRET_ACCESS_KEY=...
AWS_REGION=us-east-1
Type Imports
import type {
GenerateOptions,
GenerateResult,
StreamOptions,
StreamResult,
ChatMessage,
ToolInfo,
MCPServerInfo,
RAGConfig,
ProviderStatus,
} from "@juspay/neurolink";
Getting Help
- Check the relevant reference file above
- Review troubleshooting.md for common issues
- Look at code templates in
templates/
- Read the full CLAUDE.md in the project root for architecture details
Instructions for Claude: Based on the user's query about $ARGUMENTS, read the appropriate reference file and provide specific guidance. If no topic is specified, give a general overview of NeuroLink capabilities and ask what they'd like help with.
1---2name: neurolink-guide3description: Guide for using the NeuroLink SDK and CLI. Invoke when users ask how to use neurolink, integrate AI providers, add MCP tools, configure RAG, set up memory, deploy servers, or work with multimodal content. Covers SDK, CLI, providers, tools, and enterprise features.4---56# NeuroLink Usage Guide78NeuroLink is an enterprise AI development platform providing unified access to 30+ AI providers (text, voice, multimodal) through a single API. It ships as both a TypeScript SDK (`@juspay/neurolink`) and a professional CLI.910## Quick Navigation1112Based on your query, I'll guide you to the right documentation:1314- **Getting Started** → Read sdk-quickstart.md15- **Provider Setup** → Read providers.md16- **Multimodal (images, PDFs, files)** → Read multimodal.md17- **MCP Tools Integration** → Read tools-mcp.md18- **RAG Pipelines** → Read rag-integration.md19- **Conversation Memory** → Read memory-conversations.md20- **CLI Commands** → Read cli-reference.md21- **Advanced Features** → Read advanced-features.md22- **Troubleshooting** → Read troubleshooting.md2324## Topic Routing2526If the user asked about `$ARGUMENTS`:2728| Topic Keywords | Reference File |29| -------------------------------------------------------------------------- | ----------------------- |30| install, setup, start, begin, quickstart | sdk-quickstart.md |31| provider, openai, anthropic, vertex, bedrock, azure, gemini, claude, model | providers.md |32| image, pdf, csv, excel, document, file, multimodal, vision | multimodal.md |33| tool, mcp, server, GitHub, external, function | tools-mcp.md |34| rag, retrieval, chunk, vector, embed, document search | rag-integration.md |35| memory, conversation, history, session, redis, context | memory-conversations.md |36| cli, command, terminal, generate, stream, loop, serve | cli-reference.md |37| hitl, workflow, agent, observe, telemetry, deploy, server | advanced-features.md |38| error, issue, problem, fix, debug, not working | troubleshooting.md |3940## Installation4142```bash43npm install @juspay/neurolink44# or45pnpm add @juspay/neurolink46# or47yarn add @juspay/neurolink48```4950## Minimal Example5152```typescript53import { NeuroLink } from "@juspay/neurolink";5455const neurolink = new NeuroLink();5657// Generate text58const result = await neurolink.generate("Explain quantum computing");59console.log(result.content);6061// Stream response62const stream = await neurolink.stream({ input: { text: "Write a story" } });63for await (const chunk of stream.stream) {64 if ("content" in chunk) {65 process.stdout.write(chunk.content);66 }67}68```6970## Key Capabilities7172| Feature | Description |73| ----------------- | ---------------------------------------------------------------- |74| **30+ Providers** | OpenAI, Anthropic, Vertex, Bedrock, Azure, Mistral, Ollama, etc. |75| **Multimodal** | Images, PDFs, CSV, Excel, Word, 50+ file types |76| **MCP Tools** | 58+ tools via Model Context Protocol |77| **RAG** | Built-in chunking, embedding, vector search |78| **Memory** | Conversation history with Redis support |79| **Streaming** | Real-time token streaming |80| **HITL** | Human-in-the-loop approval workflows |81| **Observability** | Langfuse, OpenTelemetry integration |8283## Code Templates8485Ready-to-use examples in `templates/`:8687- `templates/basic-setup.ts` - Basic SDK initialization88- `templates/streaming.ts` - Streaming responses89- `templates/with-tools.ts` - Tool integration90- `templates/rag-pipeline.ts` - RAG usage91- `templates/server-deploy.ts` - HTTP server deployment9293## CLI Quick Reference9495```bash96# Generate content97neurolink generate "Your prompt"9899# Stream output100neurolink stream "Write a story"101102# Interactive mode103neurolink loop104105# Start HTTP server106neurolink serve --port 3000107108# Setup providers109neurolink setup openai110```111112## Environment Variables113114Set up your provider credentials:115116```bash117# OpenAI118OPENAI_API_KEY=sk-...119120# Anthropic121ANTHROPIC_API_KEY=sk-ant-...122123# Google AI Studio124GOOGLE_API_KEY=...125126# Vertex AI127VERTEX_PROJECT_ID=...128GOOGLE_APPLICATION_CREDENTIALS=/path/to/credentials.json129130# AWS Bedrock131AWS_ACCESS_KEY_ID=...132AWS_SECRET_ACCESS_KEY=...133AWS_REGION=us-east-1134```135136## Type Imports137138```typescript139import type {140 GenerateOptions,141 GenerateResult,142 StreamOptions,143 StreamResult,144 ChatMessage,145 ToolInfo,146 MCPServerInfo,147 RAGConfig,148 ProviderStatus,149} from "@juspay/neurolink";150```151152## Getting Help1531541. Check the relevant reference file above1552. Review troubleshooting.md for common issues1563. Look at code templates in `templates/`1574. Read the full CLAUDE.md in the project root for architecture details158159---160161**Instructions for Claude:** Based on the user's query about `$ARGUMENTS`, read the appropriate reference file and provide specific guidance. If no topic is specified, give a general overview of NeuroLink capabilities and ask what they'd like help with.
juspay