Added langGraph.js

langGraph.js/.gitignore

@@ -0,0 +1,2 @@
+node_modules/
+.env

langGraph.js/Agent-1/.env.example

@@ -0,0 +1 @@
+GROQ_API_KEY=  # "your_groq_api_key" 

langGraph.js/Agent-1/README.md

@@ -0,0 +1,49 @@
+# Minimal AI Chat Agent
+
+This project showcases a lightweight “Hello World”–style AI assistant built with [LangGraph](https://github.com/langchain-ai/langgraphjs), [LangChain.js](https://js.langchain.com/), and the [Groq](https://groq.com/) language model (`llama-3.3-70b-versatile`). It accepts user input through the terminal and returns concise, accurate responses generated by the LLM.
+
+This project provides an excellent starting point for experimenting with graph-based AI logic, custom chat flows, or Groq-powered inference.
+
+## Setup Instructions
+
+### 1. Install dependencies
+```bash
+npm install
+```
+
+### 2. Create a .env file
+Create a ```.env``` file in the root directory (or a copy of .env.example ) with your Groq API key:
+
+```ini
+GROQ_API_KEY="your_groq_api_key_here"
+```
+You can get a free API key from [Groq](https://console.groq.com).
+
+### 3. Run the Assistant
+```bash
+node index.js
+```
+(Make sure you're using Node.js version 18 or above)
+
+## Example Usage
+Once started, you can chat with the assistant in your terminal:
+
+```
+User: What are the principles of OOPS?
+AI: (whatever answer we get from LLM)
+```
+
+
+## Graph Structure
+
+Here is the visual representation of the LangGraph workflow.
+
+![Graph Structure](./assets/graph.png)
+
+## Results
+
+Here is a sample output from the terminal after interacting with the assistant:
+
+![AI Result Screenshot](./assets/result1.png)
+
+![AI Result Screenshot](./assets/result2.png)

langGraph.js/Agent-1/index.js

@@ -0,0 +1,54 @@
+import { config } from 'dotenv' // Load environment variables from .env file
+config()
+
+// Import prompt-sync for user input in terminal
+import promptSync from 'prompt-sync';
+const prompt = promptSync();
+
+// Import necessary LangChain modules
+import { ChatGroq } from "@langchain/groq";
+import { Graph} from '@langchain/langgraph'
+
+// Initialize the ChatGroq model with API key and parameters
+const model = new ChatGroq({
+  apiKey: process.env.GROQ_API_KEY, // API key stored in .env
+  model: 'llama-3.3-70b-versatile', // Selected Groq model
+  temperature: 0.1                  // Low temperature for focused responses
+})
+
+// Create a new LangGraph workflow 
+const graph = new Graph()
+
+// Add a node named "chat" that handles LLM response generation
+graph.addNode('chat', async (input) => {
+  // Invoke the model with the incoming user message
+  const res = await model.invoke([{ role: "user", content: input }]);
+  return res.content ;  // Return only the generated text
+});
+
+// Link start → chat → end in the graph
+graph.addEdge("__start__", "chat");
+graph.addEdge("chat" ,"__end__")
+
+// Compile the graph for execution
+const compiledGraph = graph.compile();
+
+// Function to interact with the agent
+const agentCall = async () => {
+  try {
+
+    // Prompt user for input
+    let userInput = prompt("Hi user! I am you AI Assistant. Enter your query: ");
+
+    // Invoke the graph with a system-style instruction + user query
+    const result = await compiledGraph.invoke("You are a helpful AI assistant. Provide a detailed yet concise response to user query in about 50 - 100 words with accuracy. The user query is: " + userInput);
+    console.log("Result: ", result); // Output the final model response
+  } 
+  catch (err) {
+    // Catch and display errors from graph execution or API calls
+    console.error("Error running graph: ", err);
+  }
+};
+
+// Start the agent interaction
+agentCall();