👻 DeepSearch

DeepSearch is an LLM (Large Language Model) powered search engine application that allows users to search for information from various sources. This project was created using a backend developed with the GoFiber framework, a Svelte-based frontend, and a PostgreSQL database. It also offers a rich user interface with modern tools like Tailwind CSS.

---

🚀 Features

• LLM-Powered Summarization and Analysis: Summarizes and analyzes search results.
• Multiple Search Engine Support: Fetches data from search engines like Google, Yandex, and Bing.
• Sqlite Database: Used to store and manage search results.
• Fiber Framework: A fast and scalable backend.
• Svelte Frontend: A user-friendly and high-performance interface.
• Tailwind CSS: A modern and stylish design.

🛠️ Installation

Requirements

• Go (v1.19+)
• Node.js (v16+)
• Sqlite (v3+)

Steps

1. Clone the Repository:

   git clone https://github.com/username/deepsearch.git
   cd deepsearch

2. Install Backend Dependencies:

   go mod tidy

3. Install Frontend Dependencies:

   cd web
   npm install

4. Set Up the Database: Create a database in PostgreSQL and update the dsn value in the config/server.ini file:

   [db]
   dsn="postgresql://user:password@localhost:5432/deepsearch"

5. Set Up Search Engine and LLM API Keys: Fill in the key and gemini fields in the config/search.ini file:

   [serpapi]
   key = "YOUR_SERPAPI_KEY"
   
   [ai]
   gemini = "YOUR_LLM_API_KEY"

6. Run the Backend:

   go run main.go

7. Run the Frontend:

   cd web
   npm run dev

8. Open the Application: Go to http://localhost:3000 in your browser.

🔧 Configuration

server.ini

Contains API and database settings:

[api]
port = ":3000"

[db]
dsn = "postgresql://user:password@localhost:5432/deepsearch"

search.ini

Contains search engine and LLM settings:

[serpapi]
key = "YOUR_SERPAPI_KEY"
google = true
yandex = true
bing = true

[ai]
gemini = "YOUR_GEMINI_API_KEY"
prompt = "Based on the data here, I want you to extract a summary and analyze it..."

📜 Usage

1. Perform a Search: Enter a query on the main page and click the "Search" button.
2. View the Results: The search results are displayed in a summarized and analyzed form.
3. Error Messages: If an error occurs, an appropriate message is shown to the user.

🤝 Contributing

If you want to contribute, please submit a pull request or open an issue. All contributions are welcome!

🌟 Support

If you like this project, please support it by giving it a ⭐! 😊