Content
# Book Fetch MCP
Talk to any published book in the world right inside claude (or your other MCP clients).
## Feature list
- [x] write mcp tool, test basic version
- [x] if book too long, maintain in MCP cache, and paginate through it
- [ ] use gemini 1.5 for asking questions from book, feed that response to claude (due to GINORMOUS context length)
- [ ] spawn mini-RAG engine inside the server, ingestion, chunking, reranking. maintained through to end of app lifecycle
- [ ] integrate scihub as well. FREE THE WORLD'S KNOWLEDGEEE
## Setup
- make sure you have uv installed
- install deps. `cd book-fetch && `
- add this to your `claude_desktop_config.json`
```json
{
"mcpServers": {
"book-fetcher": {
"command": "uv",
"args": [
"--directory",
"<PATH_TO_PARENT_DIR>/libgen-mcp",
"run",
"main.py"
]
}
}
}
```
Connection Info
You Might Also Like
MarkItDown MCP
Converting files and office documents to Markdown.
Filesystem
Model Context Protocol Servers
Sequential Thinking
Offers a structured approach to dynamic and reflective problem-solving,...
TrendRadar
🎯 Say goodbye to information overload. AI helps you understand news hotspots...
Github
GitHub's official MCP Server
opik
Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic...