Ice Breaker

Ice Breaker is a web application that crawls LinkedIn and Twitter data about a person to generate personalized ice breakers for networking and outreach.

---

Features

• Scrapes LinkedIn and Twitter profiles for public data
• Uses AI to craft custom ice breakers based on the gathered information
• Integrates with OpenAI, Proxycurl, and Twitter APIs

---

Prerequisites

Before running the project, ensure you have the following API keys and environment variables set in a .env file. To do that, simply copy the .env.example file to .env and fill in the values.

• Note:

  • This project uses paid API services:

    • Proxycurl
    • Twitter API (paid)

  • If you enable tracing (LANGCHAIN_TRACING_V2=true), you must provide a valid LANGCHAIN_API_KEY. Without it, the app will throw an error. If you don't need tracing, omit or comment out these variables.

---

Getting Started

1. Clone the Repository

git clone https://github.com/phillippbertram/langchain-icebreaker.git
cd langchain-icebreaker

2. Install uv

If you don't have uv installed, you can install it with:

pip install uv

3. Install Dependencies

uv sync

4. Start the Flask Server

uv run python app.py

5. Open the Application

Open your browser and navigate to http://localhost:5001 to access the application.

License

MIT

---