AI Newsletter Agent 📰

An intelligent agent capable of autonomously finding relevant news articles on a given topic and compiling them into a professional, engaging newsletter.

Built with FastAPI, Inngest, LangChain, and OpenAI.

🚀 Features

• Automated Web Research: Uses BrightData SERP API to fetch real-time news and "organic" search results.
• AI Content Generation: Leverages LLMs (via OpenRouter/OpenAI) to synthesize articles into a cohesive newsletter.
• Resilient Orchestration: Powered by Inngest for retries, step management, and durability.
• Asynchronous Processing: Fully async implementation for high concurrency.
• Markdown Export: auto-saves newsletters to local Markdown files.

🛠️ Tech Stack

• Language: Python 3.11+
• Framework: FastAPI
• Orchestration: Inngest
• Search: BrightData (proxies & SERP)
• LLM: OpenAI / OpenRouter
• Dependency Management: uv

📦 Installation

1. Clone the repository

   git clone <your-repo-url>
   cd newsletter_agent

2. Install dependencies Using uv (recommended):

   uv sync

   Or standard pip:

   pip install .

3. Environment Setup Create a .env file in the root directory:

   BRIGHT_DATA_API_KEY=your_key_here
   OPENAI_API_KEY=your_key_here

🏃‍♂️ Running the Agent

1. Start the Inngest Dev Server

   npx inngest-cli@latest dev

2. Start the FastAPI App

   uv run uvicorn main:app --reload

   The app will be available at http://127.0.0.1:8000

3. Trigger a Newsletter Generation You can trigger the newsletter/generate event via the Inngest dashboard or using the provided script:

   uv run run.py

📂 Project Structure

├── main.py                 # FastAPI application and Inngest entry point
├── newsletter_service.py   # Core logic for Search and AI generation
├── custom_types.py         # Pydantic models
├── prompt.py               # Prompt engineering templates
├── run.py                  # Test script to dispatch events
└── newsletters/            # Generated output directory

📝 License

MIT