NeuroLink

1<div align="center">
2  <h1>🧠 NeuroLink</h1>
3  <p><strong>The Enterprise AI SDK for Production Applications</strong></p>
4  <p>13 Providers | 58+ MCP Tools | HITL Security | Redis Persistence</p>
5</div>
6 
7<div align="center">
8 
9[![npm version](https://badge.fury.io/js/%40juspay%2Fneurolink.svg)](https://www.npmjs.com/package/@juspay/neurolink)
10[![npm downloads](https://img.shields.io/npm/dw/@juspay/neurolink)](https://www.npmjs.com/package/@juspay/neurolink)
11[![Build Status](https://github.com/juspay/neurolink/actions/workflows/ci.yml/badge.svg)](https://github.com/juspay/neurolink/actions/workflows/ci.yml)
12[![Coverage Status](https://coveralls.io/repos/github/juspay/neurolink/badge.svg?branch=main)](https://coveralls.io/github/juspay/neurolink?branch=main)
13[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
14[![TypeScript](https://img.shields.io/badge/TypeScript-5.0+-blue.svg)](https://www.typescriptlang.org/)
15[![GitHub Stars](https://img.shields.io/github/stars/juspay/neurolink)](https://github.com/juspay/neurolink/stargazers)
16[![Discord](https://img.shields.io/discord/DISCORD_SERVER_ID?label=Discord&logo=discord)](https://discord.gg/neurolink)
17 
18</div>
19 
20Enterprise AI development platform with unified provider access, production-ready tooling, and an opinionated factory architecture. NeuroLink ships as both a TypeScript SDK and a professional CLI so teams can build, operate, and iterate on AI features quickly.
21 
22## 🧠 What is NeuroLink?
23 
24**NeuroLink is the universal AI integration platform that unifies 13 major AI providers and 100+ models under one consistent API.**
25 
26Extracted from production systems at Juspay and battle-tested at enterprise scale, NeuroLink provides a production-ready solution for integrating AI into any application. Whether you're building with OpenAI, Anthropic, Google, AWS Bedrock, Azure, or any of our 13 supported providers, NeuroLink gives you a single, consistent interface that works everywhere.
27 
28**Why NeuroLink?** Switch providers with a single parameter change, leverage 64+ built-in tools and MCP servers, deploy with confidence using enterprise features like Redis memory and multi-provider failover, and optimize costs automatically with intelligent routing. Use it via our professional CLI or TypeScript SDK—whichever fits your workflow.
29 
30**Where we're headed:** We're building for the future of AI—edge-first execution and continuous streaming architectures that make AI practically free and universally available. **[Read our vision →](docs/about/vision.md)**
31 
32**[Get Started in <5 Minutes →](docs/getting-started/quick-start.md)**
33 
34---
35 
36## What's New (Q1 2026)
37 
38| Feature                             | Version | Description                                                                                                                                                   | Guide                                                                 |
39| ----------------------------------- | ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------- |
40| **Context Window Management**       | v9.2.0  | 4-stage compaction pipeline with auto-detection, budget gate at 80% usage, per-provider token estimation                                                      | [Context Compaction Guide](docs/features/context-compaction.md)       |
41| **Tool Execution Control**          | v9.3.0  | `prepareStep` and `toolChoice` support for per-step tool enforcement in multi-step agentic loops. API-level control over tool calls.                          | [API Reference](docs/api/type-aliases/GenerateOptions.md#preparestep) |
42| **File Processor System**           | v9.1.0  | 17+ file type processors with ProcessorRegistry, security sanitization, SVG text injection                                                                    | [File Processors Guide](docs/features/file-processors.md)             |
43| **RAG with generate()/stream()**    | v9.2.0  | Pass `rag: { files }` to generate/stream for automatic document chunking, embedding, and AI-powered search. 10 chunking strategies, hybrid search, reranking. | [RAG Guide](docs/features/rag.md)                                     |
44| **External TracerProvider Support** | v8.43.0 | Integrate NeuroLink with existing OpenTelemetry instrumentation. Prevents duplicate registration conflicts.                                                   | [Observability Guide](docs/features/observability.md)                 |
45| **Server Adapters**                 | v8.43.0 | Multi-framework HTTP server with Hono, Express, Fastify, Koa support. Full CLI for server management with foreground/background modes.                        | [Server Adapters Guide](docs/guides/server-adapters/index.md)         |
46| **Title Generation Events**         | v8.38.0 | Emit `conversation:titleGenerated` event when conversation title is generated. Supports custom title prompts via `NEUROLINK_TITLE_PROMPT`.                    | [Conversation Memory Guide](docs/conversation-memory.md)              |
47| **Video Generation with Veo**       | v8.32.0 | Video generation using Veo 3.1 (`veo-3.1`). Realistic video generation with many parameter options                                                            | [Video Generation Guide](docs/features/video-generation.md)           |
48| **Image Generation with Gemini**    | v8.31.0 | Native image generation using Gemini 2.0 Flash Experimental (`imagen-3.0-generate-002`). High-quality image synthesis directly from Google AI.                | [Image Generation Guide](docs/image-generation-streaming.md)          |
49| **HTTP/Streamable HTTP Transport**  | v8.29.0 | Connect to remote MCP servers via HTTP with authentication headers, automatic retry with exponential backoff, and configurable rate limiting.                 | [HTTP Transport Guide](docs/mcp-http-transport.md)                    |
50 
51- **External TracerProvider Support** – Integrate NeuroLink with applications that already have OpenTelemetry instrumentation. Supports auto-detection and manual configuration. → [Observability Guide](docs/features/observability.md)
52- **Server Adapters** – Deploy NeuroLink as an HTTP API server with your framework of choice (Hono, Express, Fastify, Koa). Full CLI support with `serve` and `server` commands for foreground/background modes, route management, and OpenAPI generation. → [Server Adapters Guide](docs/guides/server-adapters/index.md)
53- **Title Generation Events** – Emit real-time events when conversation titles are auto-generated. Listen to `conversation:titleGenerated` for session tracking. → [Conversation Memory Guide](docs/conversation-memory.md#title-generation-events)
54- **Custom Title Prompts** – Customize conversation title generation with `NEUROLINK_TITLE_PROMPT` environment variable. Use `${userMessage}` placeholder for dynamic prompts. → [Conversation Memory Guide](docs/conversation-memory.md#customizing-the-title-prompt)
55- **Video Generation** – Transform images into 8-second videos with synchronized audio using Google Veo 3.1 via Vertex AI. Supports 720p/1080p resolutions, portrait/landscape aspect ratios. → [Video Generation Guide](docs/features/video-generation.md)
56- **Image Generation** – Generate images from text prompts using Gemini models via Vertex AI or Google AI Studio. Supports streaming mode with automatic file saving. → [Image Generation Guide](docs/image-generation-streaming.md)
57- **RAG with generate()/stream()** – Just pass `rag: { files: ["./docs/guide.md"] }` to `generate()` or `stream()`. NeuroLink auto-chunks, embeds, and creates a search tool the AI can invoke. 10 chunking strategies, hybrid search, 5 reranker types. → [RAG Guide](docs/features/rag.md)
58- **HTTP/Streamable HTTP Transport for MCP** – Connect to remote MCP servers via HTTP with authentication headers, retry logic, and rate limiting. → [HTTP Transport Guide](docs/mcp-http-transport.md)
59- 🧠 **Gemini 3 Preview Support** - Full support for gemini-3-flash-preview and gemini-3-pro-preview with extended thinking capabilities
60- 🎯 **Tool Execution Control** – Use `prepareStep` to enforce specific tool calls, change the LLM models per step in multi-step agentic executions. Prevents LLMs from skipping required tools. Use `toolChoice` for static control, or `prepareStep` for dynamic per-step logic. → [GenerateOptions Reference](docs/api/type-aliases/GenerateOptions.md#preparestep)
61- **Structured Output with Zod Schemas** – Type-safe JSON generation with automatic validation using `schema` + `output.format: "json"` in `generate()`. → [Structured Output Guide](docs/features/structured-output.md)
62- **CSV File Support** – Attach CSV files to prompts for AI-powered data analysis with auto-detection. → [CSV Guide](docs/features/multimodal-chat.md#csv-file-support)
63- **PDF File Support** – Process PDF documents with native visual analysis for Vertex AI, Anthropic, Bedrock, AI Studio. → [PDF Guide](docs/features/pdf-support.md)
64- **50+ File Types** – Process Excel, Word, RTF, JSON, YAML, XML, HTML, SVG, Markdown, and 50+ code languages with intelligent content extraction. → [File Processors Guide](docs/features/file-processors.md)
65- **LiteLLM Integration** – Access 100+ AI models from all major providers through unified interface. → [Setup Guide](docs/litellm-integration.md)
66- **SageMaker Integration** – Deploy and use custom trained models on AWS infrastructure. → [Setup Guide](docs/sagemaker-integration.md)
67- **OpenRouter Integration** – Access 300+ models from OpenAI, Anthropic, Google, Meta, and more through a single unified API. → [Setup Guide](docs/getting-started/providers/openrouter.md)
68- **Human-in-the-loop workflows** – Pause generation for user approval/input before tool execution. → [HITL Guide](docs/features/hitl.md)
69- **Guardrails middleware** – Block PII, profanity, and unsafe content with built-in filtering. → [Guardrails Guide](docs/features/guardrails.md)
70- **Context summarization** – Automatic conversation compression for long-running sessions. → [Summarization Guide](docs/context-summarization.md)
71- **Redis conversation export** – Export full session history as JSON for analytics and debugging. → [History Guide](docs/features/conversation-history.md)
72 
73```typescript
74// Image Generation with Gemini (v8.31.0)
75const image = await neurolink.generateImage({
76  prompt: "A futuristic cityscape",
77  provider: "google-ai",
78  model: "imagen-3.0-generate-002",
79});
80 
81// HTTP Transport for Remote MCP (v8.29.0)
82await neurolink.addExternalMCPServer("remote-tools", {
83  transport: "http",
84  url: "https://mcp.example.com/v1",
85  headers: { Authorization: "Bearer token" },
86  retries: 3,
87  timeout: 15000,
88});
89```
90 
91---
92 
93<details>
94<summary><strong>Previous Updates (Q4 2025)</strong></summary>
95 
96- **Image Generation** – Generate images from text prompts using Gemini models via Vertex AI or Google AI Studio. → [Guide](docs/image-generation-streaming.md)
97- **Gemini 3 Preview Support** - Full support for `gemini-3-flash-preview` and `gemini-3-pro-preview` with extended thinking
98- **Structured Output with Zod Schemas** – Type-safe JSON generation with automatic validation. → [Guide](docs/features/structured-output.md)
99- **CSV & PDF File Support** – Attach CSV/PDF files to prompts with auto-detection. → [CSV](docs/features/multimodal-chat.md#csv-file-support) | [PDF](docs/features/pdf-support.md)
100- **LiteLLM & SageMaker** – Access 100+ models via LiteLLM, deploy custom models on SageMaker. → [LiteLLM](docs/litellm-integration.md) | [SageMaker](docs/sagemaker-integration.md)
101- **OpenRouter Integration** – Access 300+ models through a single unified API. → [Guide](docs/getting-started/providers/openrouter.md)
102- **HITL & Guardrails** – Human-in-the-loop approval workflows and content filtering middleware. → [HITL](docs/features/hitl.md) | [Guardrails](docs/features/guardrails.md)
103- **Redis & Context Management** – Session export, conversation history, and automatic summarization. → [History](docs/features/conversation-history.md)
104 
105</details>
106 
107## Enterprise Security: Human-in-the-Loop (HITL)
108 
109NeuroLink includes a **production-ready HITL system** for regulated industries and high-stakes AI operations:
110 
111| Capability                  | Description                                               | Use Case                                   |
112| --------------------------- | --------------------------------------------------------- | ------------------------------------------ |
113| **Tool Approval Workflows** | Require human approval before AI executes sensitive tools | Financial transactions, data modifications |
114| **Output Validation**       | Route AI outputs through human review pipelines           | Medical diagnosis, legal documents         |
115| **Confidence Thresholds**   | Automatically trigger human review below confidence level | Critical business decisions                |
116| **Complete Audit Trail**    | Full audit logging for compliance (HIPAA, SOC2, GDPR)     | Regulated industries                       |
117 
118```typescript
119import { NeuroLink } from "@juspay/neurolink";
120 
121const neurolink = new NeuroLink({
122  hitl: {
123    enabled: true,
124    requireApproval: ["writeFile", "executeCode", "sendEmail"],
125    confidenceThreshold: 0.85,
126    reviewCallback: async (action, context) => {
127      // Custom review logic - integrate with your approval system
128      return await yourApprovalSystem.requestReview(action);
129    },
130  },
131});
132 
133// AI pauses for human approval before executing sensitive tools
134const result = await neurolink.generate({
135  input: { text: "Send quarterly report to stakeholders" },
136});
137```
138 
139**[Enterprise HITL Guide](docs/features/enterprise-hitl.md)** | **[Quick Start](docs/features/hitl.md)**
140 
141## Get Started in Two Steps
142 
143```bash
144# 1. Run the interactive setup wizard (select providers, validate keys)
145pnpm dlx @juspay/neurolink setup
146 
147# 2. Start generating with automatic provider selection
148npx @juspay/neurolink generate "Write a launch plan for multimodal chat"
149```
150 
151Need a persistent workspace? Launch loop mode with `npx @juspay/neurolink loop` - [Learn more →](docs/features/cli-loop-sessions.md)
152 
153## 🌟 Complete Feature Set
154 
155NeuroLink is a comprehensive AI development platform. Every feature below is production-ready and fully documented.
156 
157### 🤖 AI Provider Integration
158 
159**13 providers unified under one API** - Switch providers with a single parameter change.
160 
161| Provider              | Models                                             | Free Tier       | Tool Support | Status        | Documentation                                                           |
162| --------------------- | -------------------------------------------------- | --------------- | ------------ | ------------- | ----------------------------------------------------------------------- |
163| **OpenAI**            | GPT-4o, GPT-4o-mini, o1                            | ❌              | ✅ Full      | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#openai)            |
164| **Anthropic**         | Claude 4.5 Opus/Sonnet/Haiku, Claude 4 Opus/Sonnet | ❌              | ✅ Full      | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#anthropic)         |
165| **Google AI Studio**  | Gemini 3 Flash/Pro, Gemini 2.5 Flash/Pro           | ✅ Free Tier    | ✅ Full      | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#google-ai)         |
166| **AWS Bedrock**       | Claude, Titan, Llama, Nova                         | ❌              | ✅ Full      | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#bedrock)           |
167| **Google Vertex**     | Gemini 3/2.5 (gemini-3-\*-preview)                 | ❌              | ✅ Full      | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#vertex)            |
168| **Azure OpenAI**      | GPT-4, GPT-4o, o1                                  | ❌              | ✅ Full      | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#azure)             |
169| **LiteLLM**           | 100+ models unified                                | Varies          | ✅ Full      | ✅ Production | [Setup Guide](docs/litellm-integration.md)                              |
170| **AWS SageMaker**     | Custom deployed models                             | ❌              | ✅ Full      | ✅ Production | [Setup Guide](docs/sagemaker-integration.md)                            |
171| **Mistral AI**        | Mistral Large, Small                               | ✅ Free Tier    | ✅ Full      | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#mistral)           |
172| **Hugging Face**      | 100,000+ models                                    | ✅ Free         | ⚠️ Partial   | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#huggingface)       |
173| **Ollama**            | Local models (Llama, Mistral)                      | ✅ Free (Local) | ⚠️ Partial   | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#ollama)            |
174| **OpenAI Compatible** | Any OpenAI-compatible endpoint                     | Varies          | ✅ Full      | ✅ Production | [Setup Guide](docs/getting-started/provider-setup.md#openai-compatible) |
175| **OpenRouter**        | 200+ Models via OpenRouter                         | Varies          | ✅ Full      | ✅ Production | [Setup Guide](docs/getting-started/providers/openrouter.md)             |
176 
177**[📖 Provider Comparison Guide](docs/reference/provider-comparison.md)** - Detailed feature matrix and selection criteria
178**[🔬 Provider Feature Compatibility](docs/reference/provider-feature-compatibility.md)** - Test-based compatibility reference for all 19 features across 13 providers
179 
180---
181 
182### 🔧 Built-in Tools & MCP Integration
183 
184**6 Core Tools** (work across all providers, zero configuration):
185 
186| Tool                 | Purpose                  | Auto-Available          | Documentation                              |
187| -------------------- | ------------------------ | ----------------------- | ------------------------------------------ |
188| `getCurrentTime`     | Real-time clock access   | ✅                      | [Tool Reference](docs/sdk/custom-tools.md) |
189| `readFile`           | File system reading      | ✅                      | [Tool Reference](docs/sdk/custom-tools.md) |
190| `writeFile`          | File system writing      | ✅                      | [Tool Reference](docs/sdk/custom-tools.md) |
191| `listDirectory`      | Directory listing        | ✅                      | [Tool Reference](docs/sdk/custom-tools.md) |
192| `calculateMath`      | Mathematical operations  | ✅                      | [Tool Reference](docs/sdk/custom-tools.md) |
193| `websearchGrounding` | Google Vertex web search | ⚠️ Requires credentials | [Tool Reference](docs/sdk/custom-tools.md) |
194 
195**58+ External MCP Servers** supported (GitHub, PostgreSQL, Google Drive, Slack, and more):
196 
197```typescript
198// stdio transport - local MCP servers via command execution
199await neurolink.addExternalMCPServer("github", {
200  command: "npx",
201  args: ["-y", "@modelcontextprotocol/server-github"],
202  transport: "stdio",
203  env: { GITHUB_TOKEN: process.env.GITHUB_TOKEN },
204});
205 
206// HTTP transport - remote MCP servers via URL
207await neurolink.addExternalMCPServer("github-copilot", {
208  transport: "http",
209  url: "https://api.githubcopilot.com/mcp",
210  headers: { Authorization: "Bearer YOUR_COPILOT_TOKEN" },
211  timeout: 15000,
212  retries: 5,
213});
214 
215// Tools automatically available to AI
216const result = await neurolink.generate({
217  input: { text: 'Create a GitHub issue titled "Bug in auth flow"' },
218});
219```
220 
221**MCP Transport Options:**
222 
223| Transport   | Use Case       | Key Features                                    |
224| ----------- | -------------- | ----------------------------------------------- |
225| `stdio`     | Local servers  | Command execution, environment variables        |
226| `http`      | Remote servers | URL-based, auth headers, retries, rate limiting |
227| `sse`       | Event streams  | Server-Sent Events, real-time updates           |
228| `websocket` | Bi-directional | Full-duplex communication                       |
229 
230**[📖 MCP Integration Guide](docs/advanced/mcp-integration.md)** - Setup external servers
231**[📖 HTTP Transport Guide](docs/mcp-http-transport.md)** - Remote MCP server configuration
232 
233---
234 
235### 💻 Developer Experience Features
236 
237**SDK-First Design** with TypeScript, IntelliSense, and type safety:
238 
239| Feature                     | Description                                                                       | Documentation                                             |
240| --------------------------- | --------------------------------------------------------------------------------- | --------------------------------------------------------- |
241| **Auto Provider Selection** | Intelligent provider fallback                                                     | [SDK Guide](docs/sdk/index.md#auto-selection)             |
242| **Streaming Responses**     | Real-time token streaming                                                         | [Streaming Guide](docs/advanced/streaming.md)             |
243| **Conversation Memory**     | Automatic context management                                                      | [Memory Guide](docs/sdk/index.md#memory)                  |
244| **Full Type Safety**        | Complete TypeScript types                                                         | [Type Reference](docs/sdk/api-reference.md)               |
245| **Error Handling**          | Graceful provider fallback                                                        | [Error Guide](docs/reference/troubleshooting.md)          |
246| **Analytics & Evaluation**  | Usage tracking, quality scores                                                    | [Analytics Guide](docs/advanced/analytics.md)             |
247| **Middleware System**       | Request/response hooks                                                            | [Middleware Guide](docs/custom-middleware-guide.md)       |
248| **Framework Integration**   | Next.js, SvelteKit, Express                                                       | [Framework Guides](docs/sdk/framework-integration.md)     |
249| **Extended Thinking**       | Native thinking/reasoning mode for Gemini 3 and Claude models                     | [Thinking Guide](docs/features/thinking-configuration.md) |
250| **RAG Document Processing** | `rag: { files }` on generate/stream with 10 chunking strategies and hybrid search | [RAG Guide](docs/features/rag.md)                         |
251 
252---
253 
254### 📁 Multimodal & File Processing
255 
256**17+ file categories supported** (50+ total file types including code languages) with intelligent content extraction and provider-agnostic processing:
257 
258| Category      | Supported Types                                            | Processing                          |
259| ------------- | ---------------------------------------------------------- | ----------------------------------- |
260| **Documents** | Excel (`.xlsx`, `.xls`), Word (`.docx`), RTF, OpenDocument | Sheet extraction, text extraction   |
261| **Data**      | JSON, YAML, XML                                            | Validation, syntax highlighting     |
262| **Markup**    | HTML, SVG, Markdown, Text                                  | OWASP-compliant sanitization        |
263| **Code**      | 50+ languages (TypeScript, Python, Java, Go, etc.)         | Language detection, syntax metadata |
264| **Config**    | `.env`, `.ini`, `.toml`, `.cfg`                            | Secure parsing                      |
265| **Media**     | Images (PNG, JPEG, WebP, GIF), PDFs, CSV                   | Provider-specific formatting        |
266 
267```typescript
268// Process any supported file type
269const result = await neurolink.generate({
270  input: {
271    text: "Analyze this data and code",
272    files: [
273      "./data.xlsx", // Excel spreadsheet
274      "./config.yaml", // YAML configuration
275      "./diagram.svg", // SVG (injected as sanitized text)
276      "./main.py", // Python source code
277    ],
278  },
279});
280 
281// CLI: Use --file for any supported type
282// neurolink generate "Analyze this" --file ./report.xlsx --file ./config.json
283```
284 
285**Key Features:**
286 
287- **ProcessorRegistry** - Priority-based processor selection with fallback
288- **OWASP Security** - HTML/SVG sanitization prevents XSS attacks
289- **Auto-detection** - FileDetector identifies file types by extension and content
290- **Provider-agnostic** - All processors work across all 13 AI providers
291 
292**[📖 File Processors Guide](docs/features/file-processors.md)** - Complete reference for all file types
293 
294---
295 
296### 🏢 Enterprise & Production Features
297 
298**Production-ready capabilities for regulated industries:**
299 
300| Feature                     | Description                        | Use Case                  | Documentation                                               |
301| --------------------------- | ---------------------------------- | ------------------------- | ----------------------------------------------------------- |
302| **Enterprise Proxy**        | Corporate proxy support            | Behind firewalls          | [Proxy Setup](docs/enterprise-proxy-setup.md)               |
303| **Redis Memory**            | Distributed conversation state     | Multi-instance deployment | [Redis Guide](docs/getting-started/provider-setup.md#redis) |
304| **Cost Optimization**       | Automatic cheapest model selection | Budget control            | [Cost Guide](docs/advanced/index.md)                        |
305| **Multi-Provider Failover** | Automatic provider switching       | High availability         | [Failover Guide](docs/advanced/index.md)                    |
306| **Telemetry & Monitoring**  | OpenTelemetry integration          | Observability             | [Telemetry Guide](docs/telemetry-guide.md)                  |
307| **Security Hardening**      | Credential management, auditing    | Compliance                | [Security Guide](docs/advanced/enterprise.md)               |
308| **Custom Model Hosting**    | SageMaker integration              | Private models            | [SageMaker Guide](docs/sagemaker-integration.md)            |
309| **Load Balancing**          | LiteLLM proxy integration          | Scale & routing           | [Load Balancing](docs/litellm-integration.md)               |
310 
311**Security & Compliance:**
312 
313- ✅ SOC2 Type II compliant deployments
314- ✅ ISO 27001 certified infrastructure compatible
315- ✅ GDPR-compliant data handling (EU providers available)
316- ✅ HIPAA compatible (with proper configuration)
317- ✅ Hardened OS verified (SELinux, AppArmor)
318- ✅ Zero credential logging
319- ✅ Encrypted configuration storage
320- ✅ Automatic context window management with 4-stage compaction pipeline and 80% budget gate
321 
322**[📖 Enterprise Deployment Guide](docs/advanced/enterprise.md)** - Complete production checklist
323 
324---
325 
326## Enterprise Persistence: Redis Memory
327 
328Production-ready distributed conversation state for multi-instance deployments:
329 
330### Capabilities
331 
332| Feature                | Description                                  | Benefit                     |
333| ---------------------- | -------------------------------------------- | --------------------------- |
334| **Distributed Memory** | Share conversation context across instances  | Horizontal scaling          |
335| **Session Export**     | Export full history as JSON                  | Analytics, debugging, audit |
336| **Auto-Detection**     | Automatic Redis discovery from environment   | Zero-config in containers   |
337| **Graceful Failover**  | Falls back to in-memory if Redis unavailable | High availability           |
338| **TTL Management**     | Configurable session expiration              | Memory management           |
339 
340### Quick Setup
341 
342```typescript
343import { NeuroLink } from "@juspay/neurolink";
344 
345// Auto-detect Redis from REDIS_URL environment variable
346const neurolink = new NeuroLink({
347  conversationMemory: {
348    enabled: true,
349    store: "redis", // Automatically uses REDIS_URL
350    ttl: 86400, // 24-hour session expiration
351  },
352});
353 
354// Or explicit configuration
355const neurolinkExplicit = new NeuroLink({
356  conversationMemory: {
357    enabled: true,
358    store: "redis",
359    redis: {
360      host: "redis.example.com",
361      port: 6379,
362      password: process.env.REDIS_PASSWORD,
363      tls: true, // Enable for production
364    },
365  },
366});
367 
368// Export conversation for analytics
369const history = await neurolink.exportConversation({ format: "json" });
370await saveToDataWarehouse(history);
371```
372 
373### Docker Quick Start
374 
375```bash
376# Start Redis
377docker run -d --name neurolink-redis -p 6379:6379 redis:7-alpine
378 
379# Configure NeuroLink
380export REDIS_URL=redis://localhost:6379
381 
382# Start your application
383node your-app.js
384```
385 
386**[Redis Setup Guide](docs/getting-started/redis-quickstart.md)** | **[Production Configuration](docs/guides/redis-configuration.md)** | **[Migration Patterns](docs/guides/redis-migration.md)**
387 
388---
389 
390### 🎨 Professional CLI
391 
392**15+ commands** for every workflow:
393 
394| Command          | Purpose                              | Example                    | Documentation                             |
395| ---------------- | ------------------------------------ | -------------------------- | ----------------------------------------- |
396| `setup`          | Interactive provider configuration   | `neurolink setup`          | [Setup Guide](docs/cli/index.md)          |
397| `generate`       | Text generation                      | `neurolink gen "Hello"`    | [Generate](docs/cli/commands.md#generate) |
398| `stream`         | Streaming generation                 | `neurolink stream "Story"` | [Stream](docs/cli/commands.md#stream)     |
399| `status`         | Provider health check                | `neurolink status`         | [Status](docs/cli/commands.md#status)     |
400| `loop`           | Interactive session                  | `neurolink loop`           | [Loop](docs/cli/commands.md#loop)         |
401| `mcp`            | MCP server management                | `neurolink mcp discover`   | [MCP CLI](docs/cli/commands.md#mcp)       |
402| `models`         | Model listing                        | `neurolink models`         | [Models](docs/cli/commands.md#models)     |
403| `eval`           | Model evaluation                     | `neurolink eval`           | [Eval](docs/cli/commands.md#eval)         |
404| `serve`          | Start HTTP server in foreground mode | `neurolink serve`          | [Serve](docs/cli/commands.md#serve)       |
405| `server start`   | Start HTTP server in background mode | `neurolink server start`   | [Server](docs/cli/commands.md#server)     |
406| `server stop`    | Stop running background server       | `neurolink server stop`    | [Server](docs/cli/commands.md#server)     |
407| `server status`  | Show server status information       | `neurolink server status`  | [Server](docs/cli/commands.md#server)     |
408| `server routes`  | List all registered API routes       | `neurolink server routes`  | [Server](docs/cli/commands.md#server)     |
409| `server config`  | View or modify server configuration  | `neurolink server config`  | [Server](docs/cli/commands.md#server)     |
410| `server openapi` | Generate OpenAPI specification       | `neurolink server openapi` | [Server](docs/cli/commands.md#server)     |
411| `rag chunk`      | Chunk documents for RAG              | `neurolink rag chunk f.md` | [RAG CLI](docs/cli/commands.md#rag)       |
412 
413**RAG flags** are available on `generate` and `stream`: `--rag-files`, `--rag-strategy`, `--rag-chunk-size`, `--rag-chunk-overlap`, `--rag-top-k`
414 
415**[📖 Complete CLI Reference](docs/cli/commands.md)** - All commands and options
416 
417---
418 
419### 🤖 GitHub Action
420 
421Run AI-powered workflows directly in GitHub Actions with 13 provider support and automatic PR/issue commenting.
422 
423```yaml
424- uses: juspay/neurolink@v1
425  with:
426    anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
427    prompt: "Review this PR for security issues and code quality"
428    post_comment: true
429```
430 
431| Feature                | Description                                                                               |
432| ---------------------- | ----------------------------------------------------------------------------------------- |
433| **Multi-Provider**     | 13 providers with unified interface                                                       |
434| **PR/Issue Comments**  | Auto-post AI responses with intelligent updates                                           |
435| **Multimodal Support** | Attach images, PDFs, CSVs, Excel, Word, JSON, YAML, XML, HTML, SVG, code files to prompts |
436| **Cost Tracking**      | Built-in analytics and quality evaluation                                                 |
437| **Extended Thinking**  | Deep reasoning with thinking tokens                                                       |
438 
439**[📖 GitHub Action Guide](docs/guides/github-action.md)** - Complete setup and examples
440 
441---
442 
443## 💰 Smart Model Selection
444 
445NeuroLink features intelligent model selection and cost optimization:
446 
447### Cost Optimization Features
448 
449- **💰 Automatic Cost Optimization**: Selects cheapest models for simple tasks
450- **🔄 LiteLLM Model Routing**: Access 100+ models with automatic load balancing
451- **🔍 Capability-Based Selection**: Find models with specific features (vision, function calling)
452- **⚡ Intelligent Fallback**: Seamless switching when providers fail
453 
454```bash
455# Cost optimization - automatically use cheapest model
456npx @juspay/neurolink generate "Hello" --optimize-cost
457 
458# LiteLLM specific model selection
459npx @juspay/neurolink generate "Complex analysis" --provider litellm --model "anthropic/claude-3-5-sonnet"
460 
461# Auto-select best available provider
462npx @juspay/neurolink generate "Write code" # Automatically chooses optimal provider
463```
464 
465## Revolutionary Interactive CLI
466 
467NeuroLink's CLI goes beyond simple commands - it's a **full AI development environment**:
468 
469### Why Interactive Mode Changes Everything
470 
471| Feature       | Traditional CLI   | NeuroLink Interactive          |
472| ------------- | ----------------- | ------------------------------ |
473| Session State | None              | Full persistence               |
474| Memory        | Per-command       | Conversation-aware             |
475| Configuration | Flags per command | `/set` persists across session |
476| Tool Testing  | Manual per tool   | Live discovery & testing       |
477| Streaming     | Optional          | Real-time default              |
478 
479### Live Demo: Development Session
480 
481```bash
482$ npx @juspay/neurolink loop --enable-conversation-memory
483 
484neurolink > /set provider vertex
485✓ provider set to vertex (Gemini 3 support enabled)
486 
487neurolink > /set model gemini-3-flash-preview
488✓ model set to gemini-3-flash-preview
489 
490neurolink > Analyze my project architecture and suggest improvements
491 
492✓ Analyzing your project structure...
493[AI provides detailed analysis, remembering context]
494 
495neurolink > Now implement the first suggestion
496[AI remembers previous context and implements suggestion]
497 
498neurolink > /mcp discover
499✓ Discovered 58 MCP tools:
500   GitHub: create_issue, list_repos, create_pr...
501   PostgreSQL: query, insert, update...
502   [full list]
503 
504neurolink > Use the GitHub tool to create an issue for this improvement
505✓ Creating issue... (requires HITL approval if configured)
506 
507neurolink > /export json > session-2026-01-01.json
508✓ Exported 15 messages to session-2026-01-01.json
509 
510neurolink > exit
511Session saved. Resume with: neurolink loop --session session-2026-01-01.json
512```
513 
514### Session Commands Reference
515 
516| Command              | Purpose                                              |
517| -------------------- | ---------------------------------------------------- |
518| `/set <key> <value>` | Persist configuration (provider, model, temperature) |
519| `/mcp discover`      | List all available MCP tools                         |
520| `/export json`       | Export conversation to JSON                          |
521| `/history`           | View conversation history                            |
522| `/clear`             | Clear context while keeping settings                 |
523 
524**[Interactive CLI Guide](docs/features/interactive-cli.md)** | **[CLI Reference](docs/cli/commands.md)**
525 
526Skip the wizard and configure manually? See [`docs/getting-started/provider-setup.md`](docs/getting-started/provider-setup.md).
527 
528## CLI & SDK Essentials
529 
530`neurolink` CLI mirrors the SDK so teams can script experiments and codify them later.
531 
532```bash
533# Discover available providers and models
534npx @juspay/neurolink status
535npx @juspay/neurolink models list --provider google-ai
536 
537# Route to a specific provider/model
538npx @juspay/neurolink generate "Summarize customer feedback" \
539  --provider azure --model gpt-4o-mini
540 
541# Turn on analytics + evaluation for observability
542npx @juspay/neurolink generate "Draft release notes" \
543  --enable-analytics --enable-evaluation --format json
544 
545# RAG: Ask questions about your docs (auto-chunks, embeds, searches)
546npx @juspay/neurolink generate "What are the key features?" \
547  --rag-files ./docs/guide.md ./docs/api.md --rag-strategy markdown
548```
549 
550```typescript
551import { NeuroLink } from "@juspay/neurolink";
552 
553const neurolink = new NeuroLink({
554  conversationMemory: {
555    enabled: true,
556    store: "redis",
557  },
558  enableOrchestration: true,
559});
560 
561const result = await neurolink.generate({
562  input: {
563    text: "Create a comprehensive analysis",
564    files: [
565      "./sales_data.csv", // Auto-detected as CSV
566      "examples/data/invoice.pdf", // Auto-detected as PDF
567      "./diagrams/architecture.png", // Auto-detected as image
568      "./report.xlsx", // Auto-detected as Excel
569      "./config.json", // Auto-detected as JSON
570      "./diagram.svg", // Auto-detected as SVG (injected as text)
571      "./app.ts", // Auto-detected as TypeScript code
572    ],
573  },
574  provider: "vertex", // PDF-capable provider (see docs/features/pdf-support.md)
575  enableEvaluation: true,
576  region: "us-east-1",
577});
578 
579console.log(result.content);
580console.log(result.evaluation?.overallScore);
581 
582// RAG: Ask questions about your documents
583const answer = await neurolink.generate({
584  prompt: "What are the main architectural decisions?",
585  rag: {
586    files: ["./docs/architecture.md", "./docs/decisions.md"],
587    strategy: "markdown",
588    topK: 5,
589  },
590});
591console.log(answer.content); // AI searches your docs and answers
592```
593 
594### Gemini 3 with Extended Thinking
595 
596```typescript
597import { NeuroLink } from "@juspay/neurolink";
598 
599const neurolink = new NeuroLink();
600 
601// Use Gemini 3 with extended thinking for complex reasoning
602const result = await neurolink.generate({
603  input: {
604    text: "Solve this step by step: What is the optimal strategy for...",
605  },
606  provider: "vertex",
607  model: "gemini-3-flash-preview",
608  thinkingLevel: "medium", // Options: "minimal", "low", "medium", "high"
609});
610 
611console.log(result.content);
612```
613 
614Full command and API breakdown lives in [`docs/cli/commands.md`](docs/cli/commands.md) and [`docs/sdk/api-reference.md`](docs/sdk/api-reference.md).
615 
616## Platform Capabilities at a Glance
617 
618| Capability               | Highlights                                                                                                               |
619| ------------------------ | ------------------------------------------------------------------------------------------------------------------------ |
620| **Provider unification** | 13+ providers with automatic fallback, cost-aware routing, provider orchestration (Q3).                                  |
621| **Multimodal pipeline**  | Stream images + CSV data + PDF documents across providers with local/remote assets. Auto-detection for mixed file types. |
622| **Quality & governance** | Auto-evaluation engine (Q3), guardrails middleware (Q4), HITL workflows (Q4), audit logging.                             |
623| **Memory & context**     | Conversation memory, Mem0 integration, Redis history export (Q4), context summarization (Q4).                            |
624| **CLI tooling**          | Loop sessions (Q3), setup wizard, config validation, Redis auto-detect, JSON output.                                     |
625| **Enterprise ops**       | Proxy support, regional routing (Q3), telemetry hooks, configuration management.                                         |
626| **Tool ecosystem**       | MCP auto discovery, HTTP/stdio/SSE/WebSocket transports, LiteLLM hub access, SageMaker custom deployment, web search.    |
627 
628## Documentation Map
629 
630| Area            | When to Use                                               | Link                                                             |
631| --------------- | --------------------------------------------------------- | ---------------------------------------------------------------- |
632| Getting started | Install, configure, run first prompt                      | [`docs/getting-started/index.md`](docs/getting-started/index.md) |
633| Feature guides  | Understand new functionality front-to-back                | [`docs/features/index.md`](docs/features/index.md)               |
634| CLI reference   | Command syntax, flags, loop sessions                      | [`docs/cli/index.md`](docs/cli/index.md)                         |
635| SDK reference   | Classes, methods, options                                 | [`docs/sdk/index.md`](docs/sdk/index.md)                         |
636| RAG             | Document chunking, hybrid search, reranking, `rag:{}` API | [`docs/features/rag.md`](docs/features/rag.md)                   |
637| Integrations    | LiteLLM, SageMaker, MCP, Mem0                             | [`docs/litellm-integration.md`](docs/litellm-integration.md)     |
638| Advanced        | Middleware, architecture, streaming patterns              | [`docs/advanced/index.md`](docs/advanced/index.md)               |
639| Cookbook        | Practical recipes for common patterns                     | [`docs/cookbook/index.md`](docs/cookbook/index.md)               |
640| Guides          | Migration, Redis, troubleshooting, provider selection     | [`docs/guides/index.md`](docs/guides/index.md)                   |
641| Operations      | Configuration, troubleshooting, provider matrix           | [`docs/reference/index.md`](docs/reference/index.md)             |
642 
643### New in 2026: Enhanced Documentation
644 
645**Enterprise Features:**
646 
647- [Enterprise HITL Guide](docs/features/enterprise-hitl.md) - Production-ready approval workflows
648- [Interactive CLI Guide](docs/features/interactive-cli.md) - AI development environment
649- [MCP Tools Showcase](docs/features/mcp-tools-showcase.md) - 58+ external tools & 6 built-in tools
650 
651**Provider Intelligence:**
652 
653- [Provider Capabilities Audit](docs/reference/provider-capabilities-audit.md) - Technical capabilities matrix
654- [Provider Selection Guide](docs/guides/provider-selection.md) - Interactive decision wizard
655- [Provider Comparison](docs/reference/provider-comparison.md) - Feature & cost comparison
656 
657**Middleware System:**
658 
659- [Middleware Architecture](docs/advanced/middleware-architecture.md) - Complete lifecycle & patterns
660- [Built-in Middleware](docs/advanced/builtin-middleware.md) - Analytics, Guardrails, Evaluation
661- [Custom Middleware Guide](docs/custom-middleware-guide.md) - Build your own
662 
663**Redis & Persistence:**
664 
665- [Redis Quick Start](docs/getting-started/redis-quickstart.md) - 5-minute setup
666- [Redis Configuration](docs/guides/redis-configuration.md) - Production-ready setup
667- [Redis Migration](docs/guides/redis-migration.md) - Migration patterns
668 
669**Migration Guides:**
670 
671- [From LangChain](docs/guides/migration/from-langchain.md) - Complete migration guide
672- [From Vercel AI SDK](docs/guides/migration/from-vercel-ai-sdk.md) - Next.js focused
673 
674**Developer Experience:**
675 
676- [Cookbook](docs/cookbook/index.md) - 10 practical recipes
677- [Troubleshooting Guide](docs/guides/troubleshooting.md) - Common issues & solutions
678 
679## Integrations
680 
681- **LiteLLM 100+ model hub** – Unified access to third-party models via LiteLLM routing. → [`docs/litellm-integration.md`](docs/litellm-integration.md)
682- **Amazon SageMaker** – Deploy and call custom endpoints directly from NeuroLink CLI/SDK. → [`docs/sagemaker-integration.md`](docs/sagemaker-integration.md)
683- **Mem0 conversational memory** – Persistent semantic memory with vector store support. → [`docs/mem0-integration.md`](docs/mem0-integration.md)
684- **Enterprise proxy & security** – Configure outbound policies and compliance posture. → [`docs/enterprise-proxy-setup.md`](docs/enterprise-proxy-setup.md)
685- **Configuration automation** – Manage environments, regions, and credentials safely. → [`docs/configuration-management.md`](docs/configuration-management.md)
686- **MCP tool ecosystem** – Auto-discover Model Context Protocol tools and extend workflows. → [`docs/advanced/mcp-integration.md`](docs/advanced/mcp-integration.md)
687- **Remote MCP via HTTP** – Connect to HTTP-based MCP servers with authentication, retries, and rate limiting. → [`docs/mcp-http-transport.md`](docs/mcp-http-transport.md)
688 
689## Contributing & Support
690 
691- Bug reports and feature requests → [GitHub Issues](https://github.com/juspay/neurolink/issues)
692- Development workflow, testing, and pull request guidelines → [`docs/development/contributing.md`](docs/development/contributing.md)
693- Documentation improvements → open a PR referencing the [documentation matrix](docs/tracking/FEATURE-DOC-MATRIX.md).
694 
695---
696 
697NeuroLink is built with ❤️ by Juspay. Contributions, questions, and production feedback are always welcome.
698