🌐 LangChain Chatbot (Translator)

This is a LangChain + OpenAI powered Translator Chatbot. It takes input in English and translates it into a target language of your choice.

The project demonstrates:

• Setting up and using LangChain with JavaScript
• Creating and using prompt templates
• Making direct LLM calls and streaming responses
• Building a foundation for an interactive CLI chatbot

---

📂 Project Structure

langchain-chatbot/
│── README.md        # Documentation
│── package.json     # Project dependencies and scripts
│── .env             # Environment variables (ignored in git)
│── .env.example     # Example environment variables file
│── .gitignore       # Ignored files and folders
│── index.js         # Entry point (chatbot runner)
│── /src
│    ├── model.js    # Model configuration (OpenAI setup)
│    ├── prompt.js   # Prompt template logic
│    └── chat.js     # Chat / translation logic

---

🚀 Installation

1. Clone the Repository

git clone <repo-url>
cd langchain-chatbot

2. Install Dependencies

npm install

Required packages:

• langchain
• @langchain/core
• @langchain/openai
• dotenv

---

🔑 Environment Variables

Setup

Copy the example file and edit with your own keys:

cp .env.example .env

.env.example

# OpenAI API Key (Required)
OPENAI_API_KEY=your-openai-api-key-here

# LangSmith tracing (Optional, set to "true" to enable)
LANGSMITH_TRACING=true

# LangSmith API Key (Optional, only if using LangSmith)
LANGSMITH_API_KEY=your-langsmith-api-key-here

⚠️ Never commit .env to git.

---

📜 File Explanations

1. src/model.js

Configures the LLM model using OpenAI:

• Loads API key via dotenv.
• Creates an instance of ChatOpenAI.
• Sets parameters (model: gpt-4o-mini, temperature: 0).
• Exports the model to be used by other files.

---

2. src/prompt.js

Defines a prompt template that guides the LLM:

• System message → “Translate the following from English into {language}.”
• User message → Inserts the actual {text} provided by the user.
• Returns a ChatPromptTemplate ready to be invoked by the model.

---

3. src/chat.js

Handles the chat/translation logic:

• translateText(language, text) → Takes input and returns a full translated response.

• translateStream(language, text) → Streams the translation token by token, like live typing.

---

4. index.js

The entry point of the project.

• Calls translateText to get a complete translation.
• Calls translateStream to demonstrate streaming mode.

Run with:

node index.js

---

🖥️ Example Output

Final: Ciao, come stai?

Streaming:
¡Buenos días, mi amigo!

---

🛠️ Upcoming Improvements

• ✅ Interactive CLI chatbot (type input, get responses in real time).
• 🌐 REST API endpoint for frontend integration.
• 💾 Conversation memory with LangChain.
• 🎤 Speech-to-text and text-to-speech support.

---

🤝 Contributing

1. Fork the repo
2. Create a feature branch (git checkout -b feature/my-feature)
3. Commit changes (git commit -m "Added my feature")
4. Push to branch (git push origin feature/my-feature)
5. Open a Pull Request

---

📜 License

This project is licensed under the MIT License.