https://github.com/ourahma/web-search-ai-agent
An AI-powered application that searches, analyzes, and recommends developer tools using web scraping and a Large Language Model (LLM).
https://github.com/ourahma/web-search-ai-agent
firecrawl firecrawl-mcp mcp-server
Last synced: about 2 months ago
JSON representation
An AI-powered application that searches, analyzes, and recommends developer tools using web scraping and a Large Language Model (LLM).
- Host: GitHub
- URL: https://github.com/ourahma/web-search-ai-agent
- Owner: ourahma
- Created: 2025-08-08T12:32:38.000Z (2 months ago)
- Default Branch: main
- Last Pushed: 2025-08-09T13:53:41.000Z (about 2 months ago)
- Last Synced: 2025-08-09T15:29:34.163Z (about 2 months ago)
- Topics: firecrawl, firecrawl-mcp, mcp-server
- Language: Python
- Homepage:
- Size: 484 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Developer Tools Research Agent
## 📌 Overview
The **Developer Tools Research Agent** is an AI-powered application that searches, analyzes, and recommends developer tools using web scraping and a Large Language Model (LLM). The system leverages a multi-step workflow to:
1. Search for relevant tools based on a query.
2. Extract tool names from relevant articles.
3. Research each tool's website for technical and business details.
4. Provide summarized recommendations.
The project is built with **LangGraph**, **LangChain**, **Firecrawl**, and a custom workflow class.
---
## 🚀 Features
* 🔍 **Tool Discovery**: Finds relevant tools from the web.
* 📄 **Content Extraction**: Scrapes official pages for details.
* 🧠 **AI-Powered Analysis**: Extracts structured data (pricing, tech stack, API availability, etc.).
* 💡 **Recommendations**: Summarizes findings for decision-making.
* 🌐 **Streamlit UI**: Interactive interface replacing the console version.
---
## 🛠️ Tech Stack
* **Python 3.9+**
* [LangGraph](https://python.langchain.com/docs/langgraph)
* [LangChain](https://www.langchain.com/)
* [Firecrawl](https://www.firecrawl.dev/)
* [Streamlit](https://streamlit.io/)
* Deep Seek-compatible LLM API (via TogetherAI)
---
## 📂 Project Structure
```
.
├── src/
│ ├── workflow.py # Main workflow logic
│ ├── firecrawl.py # Firecrawl service wrapper
│ ├── prompts.py # Prompt templates
│ └── models.py # Pydantic models for data
├── app.py # Streamlit UI
├── requirements.txt # Dependencies
├── .env # Environment variables
├── screenshots/ # Example output screenshot
└── README.md # This file
```
---
## ⚙️ Installation
```bash
git clone git@github.com:ourahma/git@github.com:ourahma/Web-Search-AI-Agent.git.git
cd Web-Search-AI-Agent
pip install -r requirements.txt
```
Create a `.env` file with your API keys:
```env
TOGETHER_API_KEY=your_api_key_here
FIRECRAWL_API_KEY=your_api_key_here
```
---
## 💻 Usage
Run the Streamlit UI:
```bash
streamlit run app.py
```
Enter a developer tools query and view structured results with recommendations.
---
## 📊 Example Results
Below is an example of the results interface:
---
## ✍️ Author
**OURAHMA Maroua**