AGI News ✨
Repository

AGI News ✨

AGI News is a daily AI newsletter that's completely sourced by autonomous AI agents. This is an open-source project built with AI Agents, Resend, and Firecrawl πŸ”₯

Research

Description

Features

  • Autonomous AI agents gather the latest AI news.
  • Daily newsletter delivery.
  • Frontend for subscribing to the newsletter.

Installation

Clone the repository:

git clone https://github.com/ericciarla/aginews.git

Frontend Setup

Navigate to the frontend directory:

cd /frontend

Install dependencies:

npm install

Create a .env file with the following variables:

SUPABASE_URL=
SUPABASE_SECRET_KEY=
RESEND_API_KEY=

Start the development server:

npm run dev

Backend Setup

Navigate to the backend directory:

cd /backend

Install dependencies:

npm install

Create a .env file with the following variables:

FIRECRAWL_API_KEY=
SUPABASE_URL=
SUPABASE_SECRET_KEY=
X_API_BEARER_TOKEN=
OPENAI_API_KEY=
RESEND_API_KEY=

Run the backend cron job:

ts-node src/index.ts

Environment Variables

Frontend

  • SUPABASE_URL
  • SUPABASE_SECRET_KEY

Backend

  • FIRECRAWL_API_KEY
  • SUPABASE_URL
  • SUPABASE_SECRET_KEY
  • X_API_BEARER_TOKEN
  • OPENAI_API_KEY
  • RESEND_API_KEY

Related Templates

Explore more templates similar to this one

Playground

Zed.dev Crawl

The first step of many to create an LLM-friendly document for Zed's configuration.

/crawl
Playground

Developers.campsite.com Crawl

/crawl
Snippet

o3 mini Company Researcher

This Python script integrates SerpAPI, OpenAI's O3 Mini model, and Firecrawl to create a comprehensive company research tool. The workflow begins by using SerpAPI to search for company information, then leverages the O3 Mini model to intelligently select the most relevant URLs from search results, and finally employs Firecrawl's extraction API to pull detailed information from those sources. The code includes robust error handling, polling mechanisms for extraction results, and clear formatting of the output, making it an efficient tool for gathering structured company information based on specific user objectives.

o3 mini
Research
Snippet

o1 Web Crawler

o1
Crawler
Playground

Docs.google.com Scrape

/scrape
Playground

test

/scrape
Snippet

Llama 4 Maverick Web Extractor

This Python script integrates SerpAPI, Together AI's Llama 4 Maverick model (specifically "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8"), and Firecrawl to extract structured company information. The workflow first uses SerpAPI to search for company data, then employs the Llama 4 model to intelligently select the most relevant URLs (prioritizing official sources and limiting to 3 URLs), and finally leverages Firecrawl's extraction API to pull detailed information from those sources. The code includes robust error handling, logging, and polling mechanisms to ensure reliable data extraction across the entire process.

Llama 4
Extractor
Snippet

Company Researcher with GPT 4.1

Search for company information with Firecrawl and GPT 4.1

/scrape