# NeuroLink Documentation (Summary) > Enterprise AI Development Platform - Unified provider access, MCP integration, professional CLI Generated: 2026-05-09T14:13:10.162Z Full documentation: https://docs.neurolink.ink/llms-full.txt --- ## Project Overview NeuroLink is an enterprise AI development platform that provides: - Unified access to 13+ AI providers through a single consistent API - 58+ MCP (Model Context Protocol) tools and integrations - TypeScript SDK and professional CLI - Production-ready features: Redis memory, failover, telemetry - Multimodal support: text, images, PDFs, CSV, audio, video ## Supported Providers - **OpenAI** (`openai`) - **Anthropic Claude** (`anthropic`) - **Google AI Studio (Gemini)** (`google-ai`) - **Google Vertex AI** (`vertex`) - **AWS Bedrock** (`bedrock`) - **Azure OpenAI** (`azure`) - **Mistral AI** (`mistral`) - **LiteLLM (100+ models)** (`litellm`) - **OpenRouter** (`openrouter`) - **Ollama (Local)** (`ollama`) - **Hugging Face** (`huggingface`) - **AWS SageMaker** (`sagemaker`) - **OpenAI-Compatible** (`openai-compatible`) ## API Signatures Summary ### Core Methods - `neurolink.generate(options)` - Generate text response - `neurolink.stream(options)` - Stream text response - `neurolink.generateImage(options)` - Generate images (Gemini/Vertex) ### Configuration - `new NeuroLink(config)` - Initialize SDK - `neurolink.addExternalMCPServer(name, config)` - Add MCP server - `neurolink.registerTool(tool)` - Register custom tool ### Common Options - `provider` - AI provider slug (openai, anthropic, google-ai, etc.) - `model` - Model name - `input.text` - Prompt text - `input.images` - Image attachments - `maxTokens` - Response length limit - `temperature` - Creativity (0-1) - `thinkingLevel` - Extended thinking (minimal, low, medium, high) - `structuredOutput` - Zod schema for typed responses ### CLI Commands - `neurolink generate ` - Generate text - `neurolink stream ` - Stream text - `neurolink loop` - Interactive session - `neurolink setup` - Configure providers - `neurolink status` - Check provider health - `neurolink mcp list` - List MCP tools --- ## Table of Contents ### Introduction - NeuroLink ### Getting Started - Getting Started - AI Provider Guides - Quick Start - Installation - Environment Variables Configuration Guide - Anthropic Provider Guide - AWS Bedrock Provider Guide - Azure OpenAI Provider Guide - ⚙️ Provider Configuration Guide - Azure Speech Services - Deepgram Provider Guide - Redis Quick Start (5 Minutes) - DeepSeek Provider Guide - ElevenLabs Provider Guide - Google AI Studio Provider Guide - Google Vertex AI Provider Guide - Hugging Face Provider Guide - LiteLLM Provider Guide - llama.cpp Provider Guide - LM Studio Provider Guide - Mistral AI Provider Guide - NVIDIA NIM Provider Guide - Ollama Provider Guide - OpenAI-Compatible Providers Guide - OpenAI TTS Provider Guide - OpenAI Provider Guide - OpenRouter Provider Guide - Amazon SageMaker Provider Guide ### SDK Reference - SDK Reference - API Reference ### CLI - CLI Command Reference ### Features - Feature Guides - Audio Input & Transcription Guide - Authentication Providers - Auto Evaluation Engine - Client SDK - AutoResearch - Autonomous AI Experiment Engine - Claude Proxy Architecture - Claude Proxy Configuration Reference - Claude Proxy Observability - Claude Proxy Troubleshooting - Claude Proxy - Claude Subscription Testing Guide - Claude Subscription Support - CLI Loop Sessions - Context Compaction - Redis Conversation History Export - Credential Validation - CSV File Support - Embeddings - Enterprise Human-in-the-Loop System - File Processors Guide - Guardrails AI Integration with Middleware - Guardrails Implementation Guide - Guardrails Middleware - Human-in-the-Loop (HITL) Workflows - Image Generation Streaming Guide - Interactive CLI - Your AI Development Environment - MCP Enhancement Architecture Diagrams - MCP Enhancements - MCP Tools Ecosystem - 58+ Integrations - Memory Guide - Multimodal Chat Experiences - Multimodal Capabilities Guide - Observability Guide - Office Documents Support - PDF File Support - Per-Request Credentials - PPT Generation - AI-Powered Presentations - Provider Fallback & Model Chains - Provider Orchestration Brain - RAG Document Processing Guide - Real-time Voice Services - Regional Streaming Controls - Speech-to-Speech Agents: Architecture and Gemini Live Integration Plan - Streaming Guide - Structured Output with Zod Schemas - TaskManager - Scheduled & Self-Running Tasks - Extended Thinking Configuration - Text-to-Speech (TTS) Integration Guide - Video Analysis - Video Director Mode – Multi-Clip Generation & Merging - Video Generation with Veo 3.1 - Real-Time Voice Agent - Streaming Voice Loop Design - Workflow Engine Guide ### Examples - Examples & Tutorials - Advanced Examples - Basic Usage Examples - Business Applications - Tool Blocking Feature Example - Use Cases --- # Introduction ## NeuroLink NeuroLink The Enterprise AI SDK for Production Applications 21+ Providers | Voice (TTS/STT/Realtime) | 58+ MCP Tools | HITL Security | Redis Persistence [[Image: npm version]](https://www.npmjs.com/package/@juspay/neurolink) [[Image: npm downloads]](https://www.npmjs.com/package/@juspay/neurolink) [[Image: Build Status]](https://github.com/juspay/neurolink/actions/workflows/ci.yml) [Content truncated - see llms-full.txt for complete documentation] --- # Getting Started ## Getting Started # Getting Started Welcome to NeuroLink! This section will help you get up and running quickly with the Enterprise AI Development Platform. ## What You'll Learn - **[Quick Start](/docs/getting-started/quick-start)** — Get NeuroLink working in under 2 minutes with basic examples for both CLI and SDK usage. [Content truncated - see llms-full.txt for complete documentation] --- ## AI Provider Guides # AI Provider Guides Complete setup guides for all supported AI providers. ## Direct AI Providers Access leading AI models directly from their creators: ### [Anthropic](/docs/getting-started/providers/anthropic) **Claude models with API key or OAuth authentication** - Claude 4.5 Opus/Sonnet/Haiku, Claude 4.0 Opus/Sonnet - API key or OAuth (Pro/Max subscription) [Content truncated - see llms-full.txt for complete documentation] --- ## Quick Start # Quick Start Get NeuroLink running in under 2 minutes with this quick start guide. ## Prerequisites - **Node.js 18+** - **npm/pnpm/yarn** package manager - **API key** for at least one AI provider (we recommend starting with Google AI Studio - it has a free tier) ## ⚡ 1-Minute Setup ### Option 1: CLI Usage (No Installation) ```bash # Set up your API key (Google AI Studio has free tier) [Content truncated - see llms-full.txt for complete documentation] --- ## Installation # Installation Complete installation guide for NeuroLink CLI and SDK across different environments. ## Choose Your Installation Method ```bash # Direct usage (recommended) npx @juspay/neurolink generate "Hello, AI" # Global installation (optional) npm install -g @juspay/neurolink neurolink generate "Hello, AI" ``` ```bash # npm npm install @juspay/neurolink # pnpm pnpm add @juspay/neurolink [Content truncated - see llms-full.txt for complete documentation] --- ## Environment Variables Configuration Guide # Environment Variables Configuration Guide This guide provides comprehensive setup instructions for all AI providers supported by NeuroLink. The CLI automatically loads environment variables from `.env` files, making configuration seamless. ## Quick Setup ### Automatic .env Loading ✨ NEW! NeuroLink CLI automatically loads environment variables from `.env` files in your project directory: [Content truncated - see llms-full.txt for complete documentation] --- ## Anthropic Provider Guide # Anthropic Provider Guide **Direct access to Claude models with flexible authentication options** -------- | -------------------------------------- | ------------------ | | **API Key** | Production, server-side apps | Pay-per-token | | **OAuth** | Personal dev with Pro/Max subscription | Subscription quota | --- ## Quick Start ### 1. Get Your API Key [Content truncated - see llms-full.txt for complete documentation] --- ## AWS Bedrock Provider Guide # AWS Bedrock Provider Guide **Enterprise AI with Claude, Nova, Llama, Mistral, DeepSeek, Qwen, and 110+ foundation models on AWS infrastructure** ----------- | ------------------------------------------------------------------------ | ---------------------------------------- | [Content truncated - see llms-full.txt for complete documentation] --- ## Azure OpenAI Provider Guide # Azure OpenAI Provider Guide **Enterprise-grade OpenAI models with Microsoft Azure infrastructure and compliance** ## Quick Start ### 1. Create Azure OpenAI Resource ```bash # Via Azure CLI az cognitiveservices account create \ --name my-openai-resource \ --resource-group my-resource-group \ --location eastus \ --kind OpenAI \ --sku S0 ``` [Content truncated - see llms-full.txt for complete documentation] --- ## ⚙️ Provider Configuration Guide # ⚙️ Provider Configuration Guide NeuroLink supports multiple AI providers with flexible authentication methods. This guide covers complete setup for all supported providers. ## Supported Providers - **OpenAI** - GPT-4o, GPT-4o-mini, GPT-4-turbo - **Amazon Bedrock** - Claude 3.7 Sonnet, Claude 3.5 Sonnet, Claude 3 Haiku - **Amazon SageMaker** - Custom models deployed on SageMaker endpoints [Content truncated - see llms-full.txt for complete documentation] --- ## Azure Speech Services # Azure Speech Services Azure Cognitive Services Speech provides both TTS and STT capabilities. ## Setup ```bash # Required environment variables AZURE_SPEECH_KEY=your-speech-key AZURE_SPEECH_REGION=eastus ``` Get credentials from: **Azure Portal** > **Cognitive Services** > **Speech** > **Keys and Endpoint** ## Usage ### Text-to-Speech ```typescript [Content truncated - see llms-full.txt for complete documentation] --- ## Deepgram Provider Guide # Deepgram Provider Guide **Fast, accurate speech-to-text with streaming, speaker diarization, and smart formatting** > **STT-only in NeuroLink** — Deepgram is registered as the STT provider id > `deepgram`. Deepgram's TTS product is not wired in NeuroLink today; for > TTS use `openai-tts`, `elevenlabs`, `azure-tts`, or `google-ai`. [Content truncated - see llms-full.txt for complete documentation] --- ## Redis Quick Start (5 Minutes) # Redis Quick Start (5 Minutes) Get Redis storage up and running with NeuroLink in under 5 minutes. ## Prerequisites - Docker installed **OR** Redis installed locally - NeuroLink SDK installed (`pnpm add @juspay/neurolink`) ## Option 1: Docker (Recommended) The fastest way to get Redis running for development and testing. ### Start Redis Container ```bash # Start Redis with persistence [Content truncated - see llms-full.txt for complete documentation] --- ## DeepSeek Provider Guide # DeepSeek Provider Guide **Text generation with DeepSeek-V3 (chat) and DeepSeek-R1 (reasoning) through a single API** ## Quick Start ### 1. Get an API Key Sign up at [https://platform.deepseek.com](https://platform.deepseek.com) and create an API key under **API Keys**. ### 2. Configure Environment Add to your `.env` file: ```bash # Required DEEPSEEK_API_KEY=sk-... [Content truncated - see llms-full.txt for complete documentation] --- ## ElevenLabs Provider Guide # ElevenLabs Provider Guide **Studio-quality, multilingual text-to-speech with dynamic voice discovery and voice cloning** -------------- | --------------------------------------------------- | | **Provider ID** | `elevenlabs` | | **API endpoint** | `https://api.elevenlabs.io/v1` | [Content truncated - see llms-full.txt for complete documentation] --- ## Google AI Studio Provider Guide # Google AI Studio Provider Guide **Direct access to Google's Gemini models with generous free tier and simple API key authentication** ## Quick Start ### 1. Get Your API Key 1. Visit [Google AI Studio](https://aistudio.google.com/) 2. Sign in with your Google account (no GCP project needed) 3. Click **Get API Key** in the top navigation 4. Click **Create API Key** [Content truncated - see llms-full.txt for complete documentation] --- ## Google Vertex AI Provider Guide # Google Vertex AI Provider Guide **Enterprise AI on Google Cloud with Claude, Gemini, and custom models** ## Quick Start ### 1. Create GCP Project ```bash # Create project gcloud projects create my-ai-project --name="My AI Project" # Set project gcloud config set project my-ai-project # Enable Vertex AI API gcloud services enable aiplatform.googleapis.com ``` ### 2. Setup Authentication [Content truncated - see llms-full.txt for complete documentation] --- ## Hugging Face Provider Guide # Hugging Face Provider Guide **Access 100,000+ open-source AI models through Hugging Face's free inference API** ## Quick Start ### 1. Get Your API Token 1. Visit [Hugging Face](https://huggingface.co/) 2. Create a free account (no credit card required) 3. Go to [Settings → Access Tokens](https://huggingface.co/settings/tokens) 4. Click "New token" 5. Give it a name (e.g., "NeuroLink") [Content truncated - see llms-full.txt for complete documentation] --- ## LiteLLM Provider Guide # LiteLLM Provider Guide **Access hundreds of AI models across 100+ providers through the NeuroLink LiteLLM provider via a LiteLLM proxy server** ## Quick Start ### 1. Set Up a LiteLLM Proxy Server [Content truncated - see llms-full.txt for complete documentation] --- ## llama.cpp Provider Guide # llama.cpp Provider Guide **Fully offline GGUF inference — connect NeuroLink directly to a `llama-server` process** ## Quick Start ### 1. Install and Build llama.cpp ```bash # Clone the repo git clone https://github.com/ggerganov/llama.cpp cd llama.cpp # Build (CPU-only — works on any machine) cmake -B build cmake --build build --config Release -j $(nproc) [Content truncated - see llms-full.txt for complete documentation] --- ## LM Studio Provider Guide # LM Studio Provider Guide **Run any GGUF model privately on your own hardware — no cloud, no API key required** ## Quick Start ### 1. Download and Start LM Studio 1. Download LM Studio from [https://lmstudio.ai](https://lmstudio.ai) for your platform. 2. Open the app and search for a model in the **Discover** tab (e.g., `Llama 3.2 3B Instruct`). [Content truncated - see llms-full.txt for complete documentation] --- ## Mistral AI Provider Guide # Mistral AI Provider Guide **European AI excellence with GDPR compliance and competitive free tier** ## Quick Start ### 1. Get Your API Key 1. Visit [Mistral AI Console](https://console.mistral.ai/) 2. Create a free account 3. Go to "API Keys" section 4. Click "Create new key" 5. Copy the key (format: `xxx...`) ### 2. Configure NeuroLink Add to your `.env` file: ```bash [Content truncated - see llms-full.txt for complete documentation] --- ## NVIDIA NIM Provider Guide # NVIDIA NIM Provider Guide **Hundreds of optimised AI models on NVIDIA's GPU-accelerated inference platform — or your own self-hosted NIM deployment** ## Quick Start ### 1. Get an API Key Sign up at [https://build.nvidia.com](https://build.nvidia.com) and create an API key under [API Keys](https://build.nvidia.com/settings/api-keys). ### 2. Configure Environment Add to your `.env` file: [Content truncated - see llms-full.txt for complete documentation] --- ## Ollama Provider Guide # Ollama Provider Guide **Run AI models locally with full privacy - no API key or cloud service required** ------------------ | ---------------------- | ------------------------------------------------- | | **Native** (default) | `/api/generate` | Standard text generation and streaming | [Content truncated - see llms-full.txt for complete documentation] --- ## OpenAI-Compatible Providers Guide # OpenAI Compatible Provider Guide **Connect to any OpenAI-compatible API: OpenRouter, vLLM, LocalAI, and more** ---------------------- | ------------------------------------ | ---------------------- | | **OpenRouter** | AI provider aggregator (100+ models) | Multi-provider access | | **vLLM** | High-performance inference server | Self-hosted models | [Content truncated - see llms-full.txt for complete documentation] --- ## OpenAI TTS Provider Guide # OpenAI TTS Provider Guide **High-quality neural text-to-speech with six distinct voices and HD quality option** ------------- | --------------------------------------------- | | **Provider ID** | `openai-tts` | | **API endpoint** | `https://api.openai.com/v1/audio/speech` | | **Models** | `tts-1` (standard), `tts-1-hd` (high quality) | [Content truncated - see llms-full.txt for complete documentation] --- ## OpenAI Provider Guide # OpenAI Provider Guide **Access GPT-5.4, GPT-5, GPT-4o, o-series reasoning models, and embedding models through the OpenAI API** ------ | --------------------------- | | `openai` | Canonical provider name | | `gpt` | Short alias for convenience | | `chatgpt` | Alternative alias | These aliases are registered in `src/lib/factories/providerRegistry.ts`. --- ## Quick Start [Content truncated - see llms-full.txt for complete documentation] --- ## OpenRouter Provider Guide # OpenRouter Provider Guide **Access 300+ AI models from 60+ providers through a single unified API** ## Quick Start ### 1. Get Your API Key Sign up at [https://openrouter.ai](https://openrouter.ai) and get your API key from [https://openrouter.ai/keys](https://openrouter.ai/keys). ### 2. Configure Environment Add your API key to `.env`: ```bash # Required OPENROUTER_API_KEY=sk-or-v1-... [Content truncated - see llms-full.txt for complete documentation] --- ## Amazon SageMaker Provider Guide # Amazon SageMaker Provider Guide **Custom model endpoints on AWS SageMaker infrastructure** > **Version:** 9.26.x | **Status:** General Availability | **Streaming:** Not Available (see warning below) ------------- | ------------- | --------------------------------------------------- | ---------------------------------- | [Content truncated - see llms-full.txt for complete documentation] --- # SDK Reference ## SDK Reference # SDK Reference The NeuroLink SDK provides a TypeScript-first programmatic interface for integrating AI capabilities into your applications. ## Overview The SDK is designed for: - **Web applications** (React, Vue, Svelte, Angular) - **Backend services** (Node.js, Express, Fastify) - **Serverless functions** (Vercel, Netlify, AWS Lambda) - **Desktop applications** (Electron, Tauri) [Content truncated - see llms-full.txt for complete documentation] --- ## API Reference This page has moved to [SDK API Reference](/docs/sdk/api-reference). --- # CLI ## CLI Command Reference # CLI Command Reference The NeuroLink CLI mirrors the SDK. Every command shares consistent options and outputs so you can prototype in the terminal and port the workflow to code later. ## Install or Run Ad-hoc ```bash # Run without installation npx @juspay/neurolink --help # Install globally npm install -g @juspay/neurolink # Local project dependency npm install @juspay/neurolink ``` [Content truncated - see llms-full.txt for complete documentation] --- # Features ## Feature Guides # Feature Guides Comprehensive guides for all NeuroLink features organized by category. Each guide includes setup, usage patterns, configuration, and troubleshooting. ------------------------------------------------------------------------ |... [Content truncated - see llms-full.txt for complete documentation] --- ## Audio Input & Transcription Guide # Audio Input & Voice Conversations Guide NeuroLink provides comprehensive audio input capabilities, enabling real-time voice conversations with AI models. This guide covers currently available features, audio specifications, and upcoming enhancements. ## Overview ### Currently Available NeuroLink supports the following audio capabilities today: [Content truncated - see llms-full.txt for complete documentation] --- ## Authentication Providers # Authentication Providers > **Status**: Stable | **Availability**: SDK + CLI + Server ## Overview NeuroLink ships with a pluggable authentication system that validates tokens, manages sessions, and enforces role-based access control (RBAC) across your AI endpoints. [Content truncated - see llms-full.txt for complete documentation] --- ## Auto Evaluation Engine # Auto Evaluation Engine NeuroLink provides an automated quality gate that scores every response using an LLM-as-judge pipeline. Scores, rationales, and severity flags are surfaced in both CLI and SDK workflows so you can monitor drift and enforce minimum quality thresholds. ## What It Does [Content truncated - see llms-full.txt for complete documentation] --- ## Client SDK # Client SDK > **Since**: v9.30.0 | **Status**: Stable | **Availability**: SDK ## Overview The NeuroLink Client SDK provides type-safe libraries for accessing NeuroLink APIs from JavaScript and TypeScript applications. It is designed for frontend apps, backend services, and full-stack frameworks alike. **Key capabilities:** [Content truncated - see llms-full.txt for complete documentation] --- ## AutoResearch - Autonomous AI Experiment Engine # AutoResearch - Autonomous AI Experiment Engine ## Overview AutoResearch is an autonomous experiment loop that proposes code changes, executes experiments, evaluates results against a deterministic metric, and keeps or discards each change — running unattended for hours. [Content truncated - see llms-full.txt for complete documentation] --- ## Claude Proxy Architecture # Claude Proxy Architecture ## 1. System Overview The Claude proxy is a local HTTP server that sits between Claude Code and the Anthropic API. It provides multi-account rotation, automatic token refresh, rate-limit handling with exponential backoff, and optional model translation to non-Anthropic providers. ### Two operational modes [Content truncated - see llms-full.txt for complete documentation] --- ## Claude Proxy Configuration Reference # Claude Proxy Configuration Reference This document is the authoritative reference for every configurable aspect of the NeuroLink Claude proxy. It covers CLI flags, the YAML config file schema, environment variables, auto-configured Claude Code settings, and all file locations. [Content truncated - see llms-full.txt for complete documentation] --- ## Claude Proxy Observability # Claude Proxy Observability This guide explains how to read the OpenObserve dashboard used to operate the NeuroLink Claude proxy. ## Source Of Truth - Dashboard definition: `docs/assets/dashboards/neurolink-proxy-observability-dashboard.json` - Live dashboard title: `NeuroLink Proxy Observability` - Default time range: `Last 30 minutes` ## First-Time Local Setup [Content truncated - see llms-full.txt for complete documentation] --- ## Claude Proxy Troubleshooting # Claude Proxy Troubleshooting This guide covers every issue encountered during development and real-world usage of the NeuroLink Claude proxy. For general proxy documentation, see [Claude Proxy](/docs/features/claude-proxy). ## Common Issues ### 1. "API Error: 400 invalid_request_error: Error" [Content truncated - see llms-full.txt for complete documentation] --- ## Claude Proxy # Claude Proxy NeuroLink includes a Claude-API-compatible proxy server that sits between Claude Code and Anthropic. It pools multiple Claude accounts, handles rate-limit failover automatically, refreshes OAuth tokens on demand before they expire, and falls back to other providers when all Claude accounts are exhausted. ## Overview ### Why use the proxy? [Content truncated - see llms-full.txt for complete documentation] --- ## Claude Subscription Testing Guide # Claude Subscription Testing Guide This document provides comprehensive testing commands and examples for the Claude subscription feature in NeuroLink, covering API key authentication, OAuth authentication for Pro/Max subscribers, subscription tier validation, and beta features. ## 1. Prerequisites ### Environment Setup Before testing, ensure you have the required environment configured: [Content truncated - see llms-full.txt for complete documentation] --- ## Claude Subscription Support # Claude Subscription Support NeuroLink provides flexible access to Anthropic's Claude models through multiple subscription tiers and authentication methods. This guide covers setup, configuration, and best practices for each tier. ## Overview Claude is available through different subscription tiers, each offering varying levels of access, rate limits, and model availability: [Content truncated - see llms-full.txt for complete documentation] --- ## CLI Loop Sessions # CLI Loop Sessions `neurolink loop` delivers a persistent CLI workspace so you can explore prompts, tweak parameters, and inspect state without restarting the CLI. Session variables, Redis-backed history, and built-in help turn the CLI into a playground for prompt engineering and operator runbooks. ## Quick Start ```bash # Enter interactive loop mode with Anthropic as the default provider [Content truncated - see llms-full.txt for complete documentation] --- ## Context Compaction # Context Compaction ## Overview NeuroLink's Context Compaction system automatically manages conversation context windows, preventing overflow errors and maintaining conversation quality as sessions grow longer. It runs transparently before every `generate()` and `stream()` call. [Content truncated - see llms-full.txt for complete documentation] --- ## Redis Conversation History Export # Redis Conversation History Export > **Since**: v7.38.0 | **Status**: Stable | **Availability**: SDK + CLI ## Overview **What it does**: Export complete conversation session history from Redis storage as JSON for analytics, debugging, and compliance auditing. [Content truncated - see llms-full.txt for complete documentation] --- ## Credential Validation Added in **v9.59.0**, NeuroLink ships a typed `ModelAccessDeniedError` and a pre-flight `sdk.checkCredentials()` API. Together they let you validate provider credentials _before_ wiring them into a long-running flow — useful for setup wizards, health checks, and surfacing actionable errors to users instead of opaque HTTP 401/403s. [Content truncated - see llms-full.txt for complete documentation] --- ## CSV File Support # CSV File Support NeuroLink provides seamless CSV file support as a **multimodal input type** - attach CSV files directly to your AI prompts for data analysis, insights, and processing. ## Overview CSV support in NeuroLink works just like image support - it's a multimodal input that gets automatically processed and injected into your prompts. The system: [Content truncated - see llms-full.txt for complete documentation] --- ## Embeddings # Embeddings > **Status**: Stable | **Availability**: SDK + CLI + Server ## Overview Embeddings convert text into dense numerical vectors that capture semantic meaning. Two texts with similar meanings produce vectors that are close together in the embedding space, enabling use cases like: - **Semantic search** -- find documents by meaning rather than exact keyword match [Content truncated - see llms-full.txt for complete documentation] --- ## Enterprise Human-in-the-Loop System # Enterprise Human-in-the-Loop System > **Since**: v7.39.0 | **Status**: Production Ready | **Availability**: SDK & CLI :::note[Feature Status - Enterprise HITL] This document describes enterprise HITL features. Some advanced features (marked as "Planned") are not yet implemented and represent the target API design for future releases. [Content truncated - see llms-full.txt for complete documentation] --- ## File Processors Guide # File Processors Guide NeuroLink includes a comprehensive file processing system that supports 20+ file types with intelligent content extraction, security sanitization, and provider-agnostic formatting. This system enables seamless multimodal AI interactions across all 13 supported providers. ## Overview The file processor system is organized into a modular architecture: ``` [Content truncated - see llms-full.txt for complete documentation] --- ## Guardrails AI Integration with Middleware # Guardrails AI Integration with Middleware This document outlines the modern, simplified approach to integrating Guardrails AI with the NeuroLink platform using the new `MiddlewareFactory`. This enhances the safety, reliability, and security of your AI applications in a modular and maintainable way. ## Overview [Content truncated - see llms-full.txt for complete documentation] --- ## Guardrails Implementation Guide # Guardrails Implementation Guide This document provides comprehensive documentation for the NeuroLink guardrails implementation, including pre-call filtering, content sanitization, and AI-powered evaluation. ## Overview The guardrails implementation provides advanced content filtering and safety mechanisms for AI interactions. It includes: [Content truncated - see llms-full.txt for complete documentation] --- ## Guardrails Middleware # Guardrails Middleware > **Since**: v7.42.0 | **Status**: Stable | **Availability**: SDK (CLI + SDK) ## Overview **What it does**: Guardrails middleware provides real-time content filtering and policy enforcement for AI model outputs, blocking profanity, PII, unsafe content, and custom-defined terms. [Content truncated - see llms-full.txt for complete documentation] --- ## Human-in-the-Loop (HITL) Workflows # Human-in-the-Loop (HITL) Workflows > **Since**: v7.39.0 | **Status**: Stable | **Availability**: SDK ## Overview **What it does**: HITL pauses AI tool execution to request explicit user approval before performing risky operations like deleting files, modifying databases, or making expensive API calls. [Content truncated - see llms-full.txt for complete documentation] --- ## Image Generation Streaming Guide # Image Generation Streaming Guide ## Overview NeuroLink supports image generation through AI models like Google Vertex AI's `gemini-3-pro-image-preview` and `gemini-2.5-flash-image`. This guide explains how image generation works in both `generate()` and `stream()` modes, including CLI usage with automatic file saving, technical architecture, and usage examples. ## Table of Contents [Content truncated - see llms-full.txt for complete documentation] --- ## Interactive CLI - Your AI Development Environment # Interactive CLI: Your AI Development Environment > **Since**: v7.0.0 | **Status**: Production Ready | **Availability**: CLI ## Why Interactive Mode? NeuroLink's Interactive CLI transforms traditional command-line usage into a persistent development environment optimized for AI workflow iteration. [Content truncated - see llms-full.txt for complete documentation] --- ## MCP Enhancement Architecture Diagrams # MCP Enhancement Architecture Diagrams Visual guides for understanding the MCP enhancement architecture, data flows, and component interactions. > **Main documentation**: For API reference, configuration options, and code examples, see [MCP Enhancements](/docs/features/mcp-enhancements). ## Overall Architecture [Content truncated - see llms-full.txt for complete documentation] --- ## MCP Enhancements # MCP Enhancements > **Since**: v9.16.0 | **Status**: Stable | **Availability**: SDK ## Overview The MCP Enhancements suite extends NeuroLink's Model Context Protocol integration with production-grade capabilities for managing tool calls at scale. These modules address the operational challenges of running multiple MCP servers in enterprise environments: [Content truncated - see llms-full.txt for complete documentation] --- ## MCP Tools Ecosystem - 58+ Integrations # MCP Tools Ecosystem: 58+ Integrations > **Since**: v7.0.0 | **Status**: Production Ready | **MCP Version**: 2024-11-05 ## Overview NeuroLink's Model Context Protocol (MCP) integration provides a **universal plugin system** that transforms the SDK from a simple AI interface into a complete AI development platform. [Content truncated - see llms-full.txt for complete documentation] --- ## Memory Guide # Memory Guide > **Since**: v9.12.0 | **Status**: Stable | **Availability**: SDK ## Overview NeuroLink includes a **memory engine** powered by the `@juspay/hippocampus` SDK. Unlike conversation memory (which tracks recent turns in a session), memory maintains a **condensed summary** of durable facts about each user across all conversations. Key characteristics: [Content truncated - see llms-full.txt for complete documentation] --- ## Multimodal Chat Experiences # Multimodal Chat Experiences NeuroLink provides full multimodal pipelines so you can mix text, URLs, and local images in a single interaction. The CLI, SDK, and loop sessions all use the same message builder, ensuring parity across workflows. ## Video Generation {#video-generation} [Content truncated - see llms-full.txt for complete documentation] --- ## Multimodal Capabilities Guide # Multimodal Capabilities Guide NeuroLink provides comprehensive multimodal support, allowing you to combine text with various media types in a single AI interaction. This guide covers all supported input types, provider capabilities, and best practices. ## Overview **Supported Input Types:** - **Images** - JPEG, PNG, GIF, WebP, HEIC (vision-capable models) [Content truncated - see llms-full.txt for complete documentation] --- ## Observability Guide # Observability Guide Enterprise-grade observability for AI operations with Langfuse and OpenTelemetry integration. ## Overview NeuroLink provides comprehensive observability features for monitoring AI operations in production: - **Langfuse Integration**: LLM-specific observability with token tracking, cost analysis, and trace visualization [Content truncated - see llms-full.txt for complete documentation] --- ## Office Documents Support # Office Documents Support NeuroLink provides seamless Office document support as a **multimodal input type** - attach DOCX, PPTX, and XLSX documents directly to your AI prompts for document analysis, data extraction, and content processing. ## Overview [Content truncated - see llms-full.txt for complete documentation] --- ## PDF File Support # PDF File Support NeuroLink provides seamless PDF file support as a **multimodal input type** - attach PDF documents directly to your AI prompts for document analysis, information extraction, and content processing. ## Overview [Content truncated - see llms-full.txt for complete documentation] --- ## Per-Request Credentials # Per-Request Credentials > **Status**: Stable | **Availability**: SDK only ## Overview NeuroLink allows provider credentials to be supplied at two levels below the environment-variable default: on the `NeuroLink` constructor (instance level) and on individual `generate()` / `stream()` calls (per-call level). Credentials are resolved in the following order of precedence: ``` [Content truncated - see llms-full.txt for complete documentation] --- ## PPT Generation - AI-Powered Presentations # PPT Generation - AI-Powered Presentations NeuroLink enables AI-powered PowerPoint presentation generation from text prompts. Transform ideas into professional, visually-appealing presentations with intelligent content planning, multiple slide types, and optional AI-generated images. ## Overview PPT generation in NeuroLink uses a multi-stage pipeline powered by any supported AI provider: [Content truncated - see llms-full.txt for complete documentation] --- ## Provider Fallback & Model Chains Added in **v9.58.0**, NeuroLink supports two complementary mechanisms for handling **model-access denial** errors at request time: the `providerFallback` callback (dynamic, code-driven) and the `modelChain` config (declarative). They give you a single place to express "if my preferred model rejects me, try this one instead" without scattering try/catch logic across every call site. [Content truncated - see llms-full.txt for complete documentation] --- ## Provider Orchestration Brain # Provider Orchestration Brain The orchestration engine introduced in 7.42.0 pairs a task classifier with a provider/model router. When enabled, NeuroLink inspects each prompt, chooses the most suitable provider/model based on capabilities and availability, and carries that preference through the fallback chain. ## Highlights [Content truncated - see llms-full.txt for complete documentation] --- ## RAG Document Processing Guide # RAG Document Processing Guide > **Since**: v8.44.0 | **Status**: Stable | **Availability**: SDK + CLI > **Provider Defaults:** When `--provider` (CLI) or `provider` (SDK) is not specified, NeuroLink defaults to **Vertex AI** with **gemini-2.5-flash** (see `src/lib/rag/ragIntegration.ts`). [Content truncated - see llms-full.txt for complete documentation] --- ## Real-time Voice Services NeuroLink integrates the two major realtime voice APIs behind a single, provider-agnostic interface: **OpenAI Realtime** (`openai-realtime`) and **Google Gemini Live** (`gemini-live`). These let you build full-duplex voice agents where audio streams in and out simultaneously, with the model responding mid-utterance and calling tools in-flight. [Content truncated - see llms-full.txt for complete documentation] --- ## Regional Streaming Controls # Regional Streaming Controls Latency, compliance, and model availability often depend on which region you call. NeuroLink threads the `region` parameter through the generate/stream stack so you can target specific data centres when working with providers that expose regional endpoints. ## Quick Start ```typescript const neurolink = new NeuroLink(); const result = await neurolink.stream({ [Content truncated - see llms-full.txt for complete documentation] --- ## Speech-to-Speech Agents: Architecture and Gemini Live Integration Plan # Speech-to-Speech Agents: Architecture and Gemini Live Integration Plan Status: Proposal (Docs only) Owner: NeuroLink Platform Last updated: 2025-09-01 ## Goals - Use `NeuroLink.stream` as the single, unified API for both text and voice streaming (no separate engine entrypoint). - Start with Google Gemini Live API (Studio) as the first realtime provider. [Content truncated - see llms-full.txt for complete documentation] --- ## Streaming Guide # Streaming Guide > **Since**: v8.0.0 | **Status**: Stable | **Availability**: SDK + CLI > > **Provider Defaults:** When `--provider` (CLI) or `provider` (SDK) is not specified, NeuroLink defaults to **Vertex AI** with **gemini-2.5-flash**. Set the `NEUROLINK_PROVIDER` or `AI_PROVIDER` environment variable to change the default provider. ## Overview [Content truncated - see llms-full.txt for complete documentation] --- ## Structured Output with Zod Schemas # Structured Output with Zod Schemas Generate type-safe, validated JSON responses using Zod schemas. Available in `generate()` function only (not `stream()`). ## Quick Start ```typescript const neurolink = new NeuroLink(); const UserSchema = z.object({ name: z.string(), age: z.number(), email: z.string(), }); const result = await neurolink.generate({ [Content truncated - see llms-full.txt for complete documentation] --- ## TaskManager - Scheduled & Self-Running Tasks # TaskManager - Scheduled & Self-Running Tasks ## Overview TaskManager adds scheduled and self-running task capabilities to NeuroLink. It enables AI agents to execute prompts on a schedule (cron, interval, or one-shot), with two execution modes: **Isolated** (fresh context per run) and **Continuation** (preserves conversation history across runs). [Content truncated - see llms-full.txt for complete documentation] --- ## Extended Thinking Configuration # Extended Thinking Configuration Enable extended thinking/reasoning modes for AI models that support deeper reasoning capabilities. This feature allows models to "think through" complex problems before providing a response. ## Overview [Content truncated - see llms-full.txt for complete documentation] --- ## Text-to-Speech (TTS) Integration Guide # Text-to-Speech (TTS) Integration Guide NeuroLink provides integrated Text-to-Speech (TTS) capabilities, allowing you to generate high-quality audio from text prompts or AI-generated responses. This feature is perfect for voice assistants, accessibility features, narration, podcasts, and more. ## Overview **Key Features:** [Content truncated - see llms-full.txt for complete documentation] --- ## Video Analysis # Video Analysis Comprehensive video analysis for NeuroLink, powered by Gemini 2.0 Flash. This feature goes beyond basic visual description—it provides a deep logical audit of video sequences to understand "why" and "how" events occur. ## Key Capabilities - **Logical Analysis**: Dissect any video to extract the underlying intent, cause-and-effect, and logical progression. [Content truncated - see llms-full.txt for complete documentation] --- ## Video Director Mode – Multi-Clip Generation & Merging # Video Director Mode Director Mode extends NeuroLink's [video generation](/docs/features/video-generation) capability to produce **multi-segment videos with seamless AI-generated transitions**. Instead of a single clip, you define an array of **segments** — each with its own prompt and image — and NeuroLink orchestrates the full pipeline: generating each clip, extracting boundary frames,... [Content truncated - see llms-full.txt for complete documentation] --- ## Video Generation with Veo 3.1 # Video Generation with Veo 3.1 NeuroLink integrates Google's Veo 3.1 model to enable AI-powered video generation with audio from image and text prompt inputs. Transform static images into dynamic, professional-quality video content with synchronized audio. ## Overview [Content truncated - see llms-full.txt for complete documentation] --- ## Real-Time Voice Agent - Streaming Voice Loop Design # Real-Time Voice Agent - Streaming Voice Loop Design **Automatic low-latency voice conversations with STT, LLM, TTS, and barge-in support** ## Problem Statement & Solution ### The Challenge Real-time voice assistants are harder than ordinary request/response chat because they must coordinate: - continuous microphone audio input - speech detection - real-time transcription [Content truncated - see llms-full.txt for complete documentation] --- ## Workflow Engine Guide # Workflow Engine Guide > **Since**: v9.20.0 | **Status**: Stable (Testing Phase) | **Availability**: SDK + CLI > > **Provider Defaults:** When `--provider` (CLI) or `provider` (SDK) is not specified, NeuroLink defaults to **Vertex AI** with **gemini-2.5-flash**. Set the `NEUROLINK_PROVIDER` or `AI_PROVIDER` environment variable to change the default provider. ## Overview [Content truncated - see llms-full.txt for complete documentation] --- # Examples ## Examples & Tutorials # Examples & Tutorials Learn NeuroLink through practical examples and step-by-step tutorials for real-world applications. ## What You'll Find Here This section contains practical implementations, use cases, and tutorials to help you integrate NeuroLink into your projects effectively. [Content truncated - see llms-full.txt for complete documentation] --- ## Advanced Examples # Advanced Examples Complex integration patterns, enterprise workflows, and sophisticated use cases for NeuroLink. ## ️ Enterprise Architecture ### Multi-Provider Load Balancing ```typescript class LoadBalancedNeuroLink { private instances: Map; private usage: Map; private limits: Map; constructor() { this.instances = new Map([ [Content truncated - see llms-full.txt for complete documentation] --- ## Basic Usage Examples # Basic Usage Examples Simple examples to get started with NeuroLink in different scenarios and programming languages. **Prerequisites**: Before running these examples, ensure you have configured at least one AI provider. See [Provider Configuration Guide](/docs/getting-started/provider-setup) for setup instructions. ## Quick Start Examples ### Simple Text Generation ```typescript [Content truncated - see llms-full.txt for complete documentation] --- ## Business Applications # Business Applications Enterprise-focused examples demonstrating NeuroLink's value in business environments, ROI optimization, and organizational workflows. ## Executive Decision Support ### Strategic Planning Assistant **Scenario**: C-level executives need AI-powered insights for strategic decisions. ```typescript class StrategyAssistant { private neurolink: NeuroLink; [Content truncated - see llms-full.txt for complete documentation] --- ## Tool Blocking Feature Example # Tool Blocking Feature Example This example demonstrates how to use the `blockedTools` feature to prevent specific tools from being executed on external MCP servers. ## Example Configuration Create or update your `.mcp-config.json` file: ```json { "mcpServers": { "filesystem": { "name": "filesystem", "command": "npx", [Content truncated - see llms-full.txt for complete documentation] --- ## Use Cases This page has moved to [Real-World Use Cases](/docs/guides/examples/use-cases). --- # Additional Documentation This is a summary version. For complete documentation, see: - Full text: https://docs.neurolink.ink/llms-full.txt - Web docs: https://docs.neurolink.ink - GitHub: https://github.com/juspay/neurolink