lit-mcp

DISCLAIMER: This project is not affiliated with or endorsed by any book websites. Use it responsibly and respect copyright laws.

lit-mcp allows LLMs to work with popular books websites (currently only Flibusta). Initially I started this project to learn some basic SpringAI and MCP concepts, but it turned out to be quite useful.

I started with Flibusta because it is one of the largest free book repositories. I haven't found any public APIs for it, however it turned out that it is pretty easy to scrape the data I needed from the website by using CSS Selectors. It also doesn't require any authentication to access its contents.

Also, I chose SpringAI and Kotlin due to these technologies being new to me, so this project served as well as a learning experience.

I was trying to use AI to generate some parts of the code especially for the parts that involve web scraping and parsing HTML. As I see it as of now, AI is quite helpful in generating boilerplate code and providing suggestions, but it still requires human supervision to ensure correctness and quality.

Although my initial goal was to check if I would be able to vibe code the whole project using AI, it turned out that this is not yet feasible for non-trivial projects and lots of human input is still required.

Features and limitations

lit-mcp supports both stdio and HTTP modes. stdio mode is useful to run the MCP server locally and connect it to LLMs that support stdio MCPs (like Claude Desktop or local LLMs that run under LMStudio).

Unfortunately not many well known LLMs support custom MCPs yet. So this project is mostly for the future when more LLMs will support custom MCPs.

Only books in Russian language are supported at the moment, since Flibusta mostly contains russian books.

Project currently supports the following set of tools:

• flibustaGetGenresList: Get all available genres list
• flibustaSearchBooksByName: Search books by name and returns their names and IDs
• flibustaGetBookInfoByIds: Get book info by book ID. Returns detailed info for each book ID such as title, authors, genres, description, download links, user rating, user reviews, etc.
• flibustaGetPopularBooksList: Get top rated books list
• flibustaGetRecommendedBooks: Get recommended books paginated (50 items per page)
• flibustaRecommendedAuthors: Get recommended authors paginated (50 items per page)

For more information, please check the source code and the tool definitions.

Usage

Please make sure you have Java installed on your system.

To build the project, you can use the following command:

make build

This will compile the source code and create a runnable JAR file in the build/libs directory.

To run the MCP server in stdio mode, you can use the following command:

make run_stdio

To run the MCP server in HTTP mode, you can use the following command:

make run_http

On Windows systems, please use gradlew.bat to build and run the project.

One of possible ways to try it out is to use Claude Desktop app. You can configure it to use lit-mcp in stdio mode by following these steps:

1. Download and install Claude Desktop from https://www.claude.com/download
2. Open Claude Desktop and go to Settings.
3. Goto Developer tab.
4. Click Edit Config button and open claude_desktop_config.json file in a text editor.
5. Add the following block to the mcpServers section:

{
  "mcpServers": {
    "lit-mcp": {
      "command": "java",
      "args": [
        "-jar",
        "/path_to_mcp_server/lit-mcp-1.0.jar",
        "--transport=stdio"
      ]
    }
  }
}

6. Save the file and restart Claude Desktop.
7. Enjoy! Try to ask Claude to search for a book or get book info, e.g. "Find me books by Isaac Asimov and provide links to download them from Flibusta".

Example output from Claude when using lit-mcp:

Contributing

Contributions are welcome!

... View full README on GitHub