@juspay/neurolink 2.0.0 → 3.0.0

package/CHANGELOG.md

@@ -1,28 +1,55 @@
-# [2.0.0](https://github.com/juspay/neurolink/compare/v1.11.3...v2.0.0) (2025-06-28)
+# [3.0.0](https://github.com/juspay/neurolink/compare/v2.1.0...v3.0.0) (2025-07-01)
 
 
 ### Features
 
-* **cli:** add command variations and stream agent support ([5fc4c26](https://github.com/juspay/neurolink/commit/5fc4c26b23bd189be52272521bdd2ca40dd55837))
+* **proxy:** add comprehensive enterprise proxy support across all providers ([9668e67](https://github.com/juspay/neurolink/commit/9668e67dfaa27831ba85d45fdf5b7739de902b28))
 
 
 ### BREAKING CHANGES
 
-* **cli:** 'generate-text' command is deprecated and will be removed in v2.0
+* **proxy:** None - fully backward compatible
 
-## [1.11.3](https://github.com/juspay/neurolink/compare/v1.11.2...v1.11.3) (2025-06-22)
+Files modified:
+- docs/ENTERPRISE-PROXY-SETUP.md (NEW) - Comprehensive enterprise proxy guide
+- docs/PROVIDER-CONFIGURATION.md - Added proxy configuration section
+- docs/CLI-GUIDE.md - Added proxy environment variables documentation
+- docs/ENVIRONMENT-VARIABLES.md - Added proxy configuration examples
+- docs/TROUBLESHOOTING.md - Added proxy troubleshooting procedures
+- .env.example - Added proxy environment variables
+- memory-bank/ - Updated with proxy implementation milestone
+- .clinerules - Added proxy success patterns
+- CHANGELOG.md - Added v2.2.0 proxy support entry
+- package.json - Updated description with enterprise features
+- README.md - Removed outdated content
+
+# [2.1.0](https://github.com/juspay/neurolink/compare/v2.0.0...v2.1.0) (2025-06-29)
+
+### Features
+
+- **timeout:** add comprehensive timeout support for all AI providers ([8610f4a](https://github.com/juspay/neurolink/commit/8610f4ade418345b0395ab72af6e675f6eec6f93))
+
+# [2.0.0](https://github.com/juspay/neurolink/compare/v1.11.3...v2.0.0) (2025-06-28)
 
+### Features
+
+- **cli:** add command variations and stream agent support ([5fc4c26](https://github.com/juspay/neurolink/commit/5fc4c26b23bd189be52272521bdd2ca40dd55837))
+
+### BREAKING CHANGES
+
+- **cli:** 'generate-text' command is deprecated and will be removed in v2.0
+
+## [1.11.3](https://github.com/juspay/neurolink/compare/v1.11.2...v1.11.3) (2025-06-22)
 
 ### Bug Fixes
 
-* resolve MCP external tools returning raw JSON instead of human-readable responses ([921a12b](https://github.com/juspay/neurolink/commit/921a12b5b31ca96bbfe3f1db05001ddb84470e14))
+- resolve MCP external tools returning raw JSON instead of human-readable responses ([921a12b](https://github.com/juspay/neurolink/commit/921a12b5b31ca96bbfe3f1db05001ddb84470e14))
 
 ## [1.11.2](https://github.com/juspay/neurolink/compare/v1.11.1...v1.11.2) (2025-06-22)
 
-
 ### Bug Fixes
 
-* **ci:** refactor auto-converted Node.js scripts ([4088888](https://github.com/juspay/neurolink/commit/408888863f8223e64269423412f5c79a35ddfe36))
+- **ci:** refactor auto-converted Node.js scripts ([4088888](https://github.com/juspay/neurolink/commit/408888863f8223e64269423412f5c79a35ddfe36))
 
 ## [1.11.1](https://github.com/juspay/neurolink/compare/v1.11.0...v1.11.1) (2025-06-21)
 

package/README.md

@@ -11,28 +11,6 @@
 
 **NeuroLink** unifies OpenAI, Bedrock, Vertex AI, Google AI Studio, Anthropic, Azure OpenAI, Hugging Face, Ollama, and Mistral AI with intelligent fallback and streaming support. Available as both a **programmatic SDK** and **professional CLI tool**. Extracted from production use at Juspay.
 
-## 🔥 **Latest Breakthrough: Full MCP Tool Integration Operational** (June 21, 2025)
-
-**MAJOR SUCCESS**: All blocking TypeScript compilation errors resolved + Complete CLI MCP integration achieved!
-
-✅ **Function Calling Ready**: AI can now execute real filesystem operations, data analysis, and system commands
-✅ **Production Validated**: 23,230+ token MCP context loading confirmed via comprehensive CLI testing
-✅ **Zero Build Errors**: Clean TypeScript compilation after resolving all 13 blocking errors
-✅ **CLI Tool Integration**: Both `generate`/`gen` and `agent-generate` commands use full MCP capabilities
-✅ **Backward Compatible**: Tools enabled by default with opt-out flag for traditional usage
-
-```bash
-# NEW: AI can now access your filesystem and execute tools (use preferred commands)
-npx @juspay/neurolink generate "List files in this directory" --provider google-ai
-
-# Alternative shorter command
-npx @juspay/neurolink gen "List files in this directory" --provider google-ai
-
-# ⚠️ DEPRECATED: generate-text will be removed in v2.0 (use 'generate' or 'gen' instead)
-# This command shows a deprecation warning and is kept for backward compatibility only
-npx @juspay/neurolink generate-text "List files in this directory" --provider google-ai
-```
-
 ## 🚀 Quick Start
 
 ### Install & Run (2 minutes)
@@ -61,6 +39,7 @@ import { createBestAIProvider } from "@juspay/neurolink";
 const provider = createBestAIProvider();
 const result = await provider.generateText({
   prompt: "Write a haiku about programming",
+  timeout: "30s", // Optional: Set custom timeout (default: 30s)
 });
 
 console.log(result.text);
@@ -100,14 +79,14 @@ npx @juspay/neurolink status
 
 ## 🛠️ MCP Integration Status (v1.11.1) ✅ **PRODUCTION READY**
 
-| Component           | Status             | Description                                         |
-| ------------------- | ------------------ | --------------------------------------------------- |
-| Built-in Tools      | ✅ **Working**     | Time tool, utilities - fully functional             |
-| External Discovery  | ✅ **Working**     | 58+ MCP servers auto-discovered from all AI tools   |
-| Tool Execution      | ✅ **Working**     | Real-time AI tool calling with built-in tools       |
+| Component           | Status             | Description                                                |
+| ------------------- | ------------------ | ---------------------------------------------------------- |
+| Built-in Tools      | ✅ **Working**     | Time tool, utilities - fully functional                    |
+| External Discovery  | ✅ **Working**     | 58+ MCP servers auto-discovered from all AI tools          |
+| Tool Execution      | ✅ **Working**     | Real-time AI tool calling with built-in tools              |
 | **External Tools**  | ✅ **SOLVED**      | **Two-step tool calling fixed - human-readable responses** |
-| **CLI Integration** | ✅ **READY**       | **Production-ready AI assistant with external tools** |
-| External Activation | 🔧 **Development** | Discovery complete, activation protocol in progress |
+| **CLI Integration** | ✅ **READY**       | **Production-ready AI assistant with external tools**      |
+| External Activation | 🔧 **Development** | Discovery complete, activation protocol in progress        |
 
 ### ✅ Quick MCP Test (v1.7.1)
 
@@ -182,36 +161,61 @@ npx @juspay/neurolink gen "What time is it?"
 # Disable tools for training-data-only responses
 npx @juspay/neurolink generate "What time is it?" --disable-tools
 
+# With custom timeout for complex prompts
+npx @juspay/neurolink generate "Explain quantum computing in detail" --timeout 1m
+
 # Real-time streaming with agent support (default)
 npx @juspay/neurolink stream "What time is it?"
 
 # Streaming without tools (traditional mode)
 npx @juspay/neurolink stream "Tell me a story" --disable-tools
 
+# Streaming with extended timeout
+npx @juspay/neurolink stream "Write a long story" --timeout 5m
+
 # Provider diagnostics
 npx @juspay/neurolink status --verbose
 
 # Batch processing
 echo -e "Write a haiku\nExplain gravity" > prompts.txt
 npx @juspay/neurolink batch prompts.txt --output results.json
+
+# Batch with custom timeout per request
+npx @juspay/neurolink batch prompts.txt --timeout 45s --output results.json
 ```
 
 ### SDK Integration
 
 ```typescript
-// SvelteKit API route
+// SvelteKit API route with timeout handling
 export const POST: RequestHandler = async ({ request }) => {
   const { message } = await request.json();
   const provider = createBestAIProvider();
-  const result = await provider.streamText({ prompt: message });
-  return new Response(result.toReadableStream());
+
+  try {
+    const result = await provider.streamText({
+      prompt: message,
+      timeout: "2m", // 2 minutes for streaming
+    });
+    return new Response(result.toReadableStream());
+  } catch (error) {
+    if (error.name === "TimeoutError") {
+      return new Response("Request timed out", { status: 408 });
+    }
+    throw error;
+  }
 };
 
-// Next.js API route
+// Next.js API route with timeout
 export async function POST(request: NextRequest) {
   const { prompt } = await request.json();
   const provider = createBestAIProvider();
-  const result = await provider.generateText({ prompt });
+
+  const result = await provider.generateText({
+    prompt,
+    timeout: process.env.AI_TIMEOUT || "30s", // Configurable timeout
+  });
+
   return NextResponse.json({ text: result.text });
 }
 ```

package/dist/cli/commands/config.d.ts

@@ -117,12 +117,12 @@ declare const ConfigSchema: z.ZodObject<{
             model: z.ZodDefault<z.ZodString>;
             timeout: z.ZodDefault<z.ZodNumber>;
         }, "strip", z.ZodTypeAny, {
-            model: string;
             timeout: number;
+            model: string;
             baseUrl: string;
         }, {
-            model?: string | undefined;
             timeout?: number | undefined;
+            model?: string | undefined;
             baseUrl?: string | undefined;
         }>>;
         mistral: z.ZodOptional<z.ZodObject<{
@@ -176,8 +176,8 @@ declare const ConfigSchema: z.ZodObject<{
             apiKey?: string | undefined;
         } | undefined;
         ollama?: {
-            model: string;
             timeout: number;
+            model: string;
             baseUrl: string;
         } | undefined;
         mistral?: {
@@ -225,8 +225,8 @@ declare const ConfigSchema: z.ZodObject<{
             apiKey?: string | undefined;
         } | undefined;
         ollama?: {
-            model?: string | undefined;
             timeout?: number | undefined;
+            model?: string | undefined;
             baseUrl?: string | undefined;
         } | undefined;
         mistral?: {
@@ -299,8 +299,8 @@ declare const ConfigSchema: z.ZodObject<{
             apiKey?: string | undefined;
         } | undefined;
         ollama?: {
-            model: string;
             timeout: number;
+            model: string;
             baseUrl: string;
         } | undefined;
         mistral?: {
@@ -360,8 +360,8 @@ declare const ConfigSchema: z.ZodObject<{
             apiKey?: string | undefined;
         } | undefined;
         ollama?: {
-            model?: string | undefined;
             timeout?: number | undefined;
+            model?: string | undefined;
             baseUrl?: string | undefined;
         } | undefined;
         mistral?: {

package/dist/cli/index.js

@@ -253,9 +253,9 @@ const cli = yargs(args)
     description: "Enable debug mode with verbose output",
 }) // Kept for potential specific debug logic
     .option("timeout", {
-    type: "number",
-    default: 30000,
-    description: "Timeout for the request in milliseconds",
+    type: "string",
+    default: "30s",
+    description: "Timeout for the request (e.g., 30s, 2m, 1h, 5000)",
 })
     .option("disable-tools", {
     type: "boolean",
@@ -268,7 +268,7 @@ const cli = yargs(args)
     .example('$0 generate-text "Hello world" --disable-tools', "Use without tool integration"), async (argv) => {
     // Check if generate-text was used specifically (for deprecation warning)
     const usedCommand = argv._[0];
-    if (usedCommand === 'generate-text' && !argv.quiet) {
+    if (usedCommand === "generate-text" && !argv.quiet) {
         console.warn(chalk.yellow('⚠️  Warning: "generate-text" is deprecated. Use "generate" or "gen" instead for multimodal support.'));
     }
     let originalConsole = {};
@@ -285,9 +285,8 @@ const cli = yargs(args)
         ? null
         : ora("🤖 Generating text...").start();
     try {
-        const timeoutPromise = new Promise((_, reject) => {
-            setTimeout(() => reject(new Error(`Request timeout (${argv.timeout}ms)`)), argv.timeout);
-        });
+        // The SDK will handle the timeout internally, so we don't need this wrapper anymore
+        // Just pass the timeout to the SDK
         // Use AgentEnhancedProvider when tools are enabled, otherwise use standard SDK
         let generatePromise;
         if (argv.disableTools === true) {
@@ -300,6 +299,7 @@ const cli = yargs(args)
                 temperature: argv.temperature,
                 maxTokens: argv.maxTokens,
                 systemPrompt: argv.system,
+                timeout: argv.timeout,
             });
         }
         else {
@@ -323,10 +323,7 @@ const cli = yargs(args)
             });
             generatePromise = agentProvider.generateText(argv.prompt);
         }
-        const result = (await Promise.race([
-            generatePromise,
-            timeoutPromise,
-        ]));
+        const result = await generatePromise;
         if (argv.format === "json" && originalConsole.log) {
             Object.assign(console, originalConsole);
         }
@@ -334,20 +331,26 @@ const cli = yargs(args)
             spinner.succeed(chalk.green("✅ Text generated successfully!"));
         }
         // Handle both AgentEnhancedProvider (AI SDK) and standard NeuroLink SDK responses
-        const responseText = result.text || result.content || "";
-        const responseUsage = result.usage || {
-            promptTokens: 0,
-            completionTokens: 0,
-            totalTokens: 0,
-        };
+        const responseText = result
+            ? result.text || result.content || ""
+            : "";
+        const responseUsage = result
+            ? result.usage || {
+                promptTokens: 0,
+                completionTokens: 0,
+                totalTokens: 0,
+            }
+            : { promptTokens: 0, completionTokens: 0, totalTokens: 0 };
         if (argv.format === "json") {
             const jsonOutput = {
                 content: responseText,
-                provider: result.provider || argv.provider,
+                provider: result
+                    ? result.provider || argv.provider
+                    : argv.provider,
                 usage: responseUsage,
-                responseTime: result.responseTime || 0,
-                toolCalls: result.toolCalls || [],
-                toolResults: result.toolResults || [],
+                responseTime: result ? result.responseTime || 0 : 0,
+                toolCalls: result ? result.toolCalls || [] : [],
+                toolResults: result ? result.toolResults || [] : [],
             };
             process.stdout.write(JSON.stringify(jsonOutput, null, 2) + "\n");
         }
@@ -357,7 +360,9 @@ const cli = yargs(args)
                 console.log("\n" + responseText + "\n");
             }
             // Show tool calls if any
-            if (result.toolCalls && result.toolCalls.length > 0) {
+            if (result &&
+                result.toolCalls &&
+                result.toolCalls.length > 0) {
                 console.log(chalk.blue("🔧 Tools Called:"));
                 for (const toolCall of result.toolCalls) {
                     console.log(`- ${toolCall.toolName}`);
@@ -366,7 +371,9 @@ const cli = yargs(args)
                 console.log();
             }
             // Show tool results if any
-            if (result.toolResults && result.toolResults.length > 0) {
+            if (result &&
+                result.toolResults &&
+                result.toolResults.length > 0) {
                 console.log(chalk.blue("📋 Tool Results:"));
                 for (const toolResult of result.toolResults) {
                     console.log(`- ${toolResult.toolCallId}`);
@@ -375,9 +382,11 @@ const cli = yargs(args)
                 console.log();
             }
             console.log(JSON.stringify({
-                provider: result.provider || argv.provider,
+                provider: result
+                    ? result.provider || argv.provider
+                    : argv.provider,
                 usage: responseUsage,
-                responseTime: result.responseTime || 0,
+                responseTime: result ? result.responseTime || 0 : 0,
             }, null, 2));
             if (responseUsage.totalTokens) {
                 console.log(chalk.blue(`ℹ️  ${responseUsage.totalTokens} tokens used`));
@@ -436,6 +445,11 @@ const cli = yargs(args)
     type: "number",
     default: 0.7,
     description: "Creativity level",
+})
+    .option("timeout", {
+    type: "string",
+    default: "2m",
+    description: "Timeout for streaming (e.g., 30s, 2m, 1h)",
 })
     .option("debug", {
     type: "boolean",
@@ -468,6 +482,7 @@ const cli = yargs(args)
                     ? undefined
                     : argv.provider,
                 temperature: argv.temperature,
+                timeout: argv.timeout,
             });
         }
         else {
@@ -498,7 +513,7 @@ const cli = yargs(args)
             const DELAY_MS = 50;
             for (let i = 0; i < text.length; i += CHUNK_SIZE) {
                 process.stdout.write(text.slice(i, i + CHUNK_SIZE));
-                await new Promise(resolve => setTimeout(resolve, DELAY_MS)); // Small delay
+                await new Promise((resolve) => setTimeout(resolve, DELAY_MS)); // Small delay
             }
             if (!argv.quiet) {
                 process.stdout.write("\n");
@@ -552,9 +567,9 @@ const cli = yargs(args)
     description: "AI provider to use",
 })
     .option("timeout", {
-    type: "number",
-    default: 30000,
-    description: "Timeout for each request in milliseconds",
+    type: "string",
+    default: "30s",
+    description: "Timeout for each request (e.g., 30s, 2m, 1h)",
 })
     .option("temperature", {
     type: "number",
@@ -607,8 +622,7 @@ const cli = yargs(args)
                 spinner.text = `Processing ${i + 1}/${prompts.length}: ${prompts[i].substring(0, 30)}...`;
             }
             try {
-                const timeoutPromise = new Promise((_, reject) => setTimeout(() => reject(new Error("Request timeout")), argv.timeout));
-                const generatePromise = sdk.generateText({
+                const result = await sdk.generateText({
                     prompt: prompts[i],
                     provider: argv.provider === "auto"
                         ? undefined
@@ -616,11 +630,8 @@ const cli = yargs(args)
                     temperature: argv.temperature,
                     maxTokens: argv.maxTokens,
                     systemPrompt: argv.system,
+                    timeout: argv.timeout,
                 });
-                const result = (await Promise.race([
-                    generatePromise,
-                    timeoutPromise,
-                ]));
                 results.push({ prompt: prompts[i], response: result.content });
                 if (spinner) {
                     spinner.render();

package/dist/core/types.d.ts

@@ -80,6 +80,7 @@ export interface TextGenerationOptions {
     systemPrompt?: string;
     schema?: ZodType<unknown, ZodTypeDef, unknown> | Schema<unknown>;
     tools?: Record<string, Tool>;
+    timeout?: number | string;
 }
 /**
  * Stream text options interface
@@ -92,6 +93,7 @@ export interface StreamTextOptions {
     systemPrompt?: string;
     schema?: ZodType<unknown, ZodTypeDef, unknown> | Schema<unknown>;
     tools?: Record<string, Tool>;
+    timeout?: number | string;
 }
 /**
  * AI Provider interface with flexible parameter support

package/dist/lib/core/types.d.ts

@@ -80,6 +80,7 @@ export interface TextGenerationOptions {
     systemPrompt?: string;
     schema?: ZodType<unknown, ZodTypeDef, unknown> | Schema<unknown>;
     tools?: Record<string, Tool>;
+    timeout?: number | string;
 }
 /**
  * Stream text options interface
@@ -92,6 +93,7 @@ export interface StreamTextOptions {
     systemPrompt?: string;
     schema?: ZodType<unknown, ZodTypeDef, unknown> | Schema<unknown>;
     tools?: Record<string, Tool>;
+    timeout?: number | string;
 }
 /**
  * AI Provider interface with flexible parameter support

package/dist/lib/mcp/plugins/filesystem-mcp.d.ts

@@ -3,7 +3,7 @@
  * Implements the new MCP contract with security sandbox
  * Based on research document recommendations
  */
-import { MCP, MCPMetadata, ExecutionContext } from "../contracts/mcp-contract.js";
+import { MCP, type MCPMetadata, type ExecutionContext } from "../contracts/mcp-contract.js";
 /**
  * FileSystem MCP Configuration
  */

package/dist/lib/neurolink.d.ts

@@ -12,6 +12,7 @@ export interface TextGenerationOptions {
     maxTokens?: number;
     systemPrompt?: string;
     schema?: any;
+    timeout?: number | string;
     disableTools?: boolean;
 }
 export interface StreamTextOptions {
@@ -20,6 +21,7 @@ export interface StreamTextOptions {
     temperature?: number;
     maxTokens?: number;
     systemPrompt?: string;
+    timeout?: number | string;
 }
 export interface TextGenerationResult {
     content: string;

package/dist/lib/neurolink.js

@@ -11,6 +11,7 @@ import { toolRegistry } from "./mcp/tool-registry.js";
 import { unifiedRegistry } from "./mcp/unified-registry.js";
 import { logger } from "./utils/logger.js";
 import { getBestProvider } from "./utils/providerUtils-fixed.js";
+import { TimeoutError } from "./utils/timeout.js";
 export class NeuroLink {
     mcpInitialized = false;
     contextManager;
@@ -32,10 +33,9 @@ export class NeuroLink {
             const mcpInitPromise = Promise.race([
                 this.doIsolatedMCPInitialization(),
                 new Promise((_, reject) => {
-                    const timer = setTimeout(() => {
+                    setTimeout(() => {
                         reject(new Error("MCP initialization timeout after 3s"));
                     }, initTimeout);
-                    timer.unref(); // Don't keep process alive
                 }),
             ]);
             await mcpInitPromise;
@@ -134,6 +134,7 @@ export class NeuroLink {
                 temperature: options.temperature,
                 maxTokens: options.maxTokens,
                 systemPrompt: enhancedSystemPrompt,
+                timeout: options.timeout,
             }, options.schema);
             if (!result) {
                 throw new Error("No response received from AI provider");
@@ -222,6 +223,7 @@ export class NeuroLink {
                     temperature: options.temperature,
                     maxTokens: options.maxTokens,
                     systemPrompt: options.systemPrompt,
+                    timeout: options.timeout,
                 }, options.schema);
                 if (!result) {
                     throw new Error("No response received from AI provider");
@@ -252,9 +254,18 @@ export class NeuroLink {
             catch (error) {
                 const errorMessage = error instanceof Error ? error.message : String(error);
                 lastError = error instanceof Error ? error : new Error(errorMessage);
+                // Special handling for timeout errors
+                if (error instanceof TimeoutError) {
+                    logger.warn(`[${functionTag}] Provider timed out`, {
+                        provider: providerName,
+                        timeout: error.timeout,
+                        operation: error.operation,
+                    });
+                }
                 logger.debug(`[${functionTag}] Provider failed, trying next`, {
                     provider: providerName,
                     error: errorMessage,
+                    isTimeout: error instanceof TimeoutError,
                     remainingProviders: tryProviders.slice(tryProviders.indexOf(providerName) + 1),
                 });
                 // Continue to next provider
@@ -338,6 +349,7 @@ Note: Tool integration is currently in development. Please provide helpful respo
                     temperature: options.temperature,
                     maxTokens: options.maxTokens,
                     systemPrompt: options.systemPrompt,
+                    timeout: options.timeout,
                 });
                 if (!result) {
                     throw new Error("No stream response received from AI provider");
@@ -358,9 +370,18 @@ Note: Tool integration is currently in development. Please provide helpful respo
             catch (error) {
                 const errorMessage = error instanceof Error ? error.message : String(error);
                 lastError = error instanceof Error ? error : new Error(errorMessage);
+                // Special handling for timeout errors
+                if (error instanceof TimeoutError) {
+                    logger.warn(`[${functionTag}] Provider timed out`, {
+                        provider: providerName,
+                        timeout: error.timeout,
+                        operation: error.operation,
+                    });
+                }
                 logger.debug(`[${functionTag}] Provider failed, trying next`, {
                     provider: providerName,
                     error: errorMessage,
+                    isTimeout: error instanceof TimeoutError,
                     remainingProviders: tryProviders.slice(tryProviders.indexOf(providerName) + 1),
                 });
                 // Continue to next provider

package/dist/lib/providers/agent-enhanced-provider.d.ts

@@ -15,6 +15,7 @@ interface AgentConfig {
     enableTools?: boolean;
     enableMCP?: boolean;
     mcpInitTimeoutMs?: number;
+    toolExecutionTimeout?: number | string;
     mcpDiscoveryOptions?: {
         searchPaths?: string[];
         configFiles?: string[];