MCP Server Librarian

🚀 Librarian

Librarian is a powerful knowledge management system designed to work seamlessly with Large Language Models (LLMs), offering comprehensive solutions for knowledge organization and retrieval.

🚀 Quick Start

Librarian is a robust knowledge management system primarily designed to work in tandem with Large Language Models (LLMs). It offers the following core capabilities:

1. Document Management: Handle stored knowledge content by fetching, listing, and searching for documents.
2. Tag Management: Manage and query classification tags for knowledge content.
3. Deep Integration: Collaborate seamlessly with LLMs via the Model Context Protocol (MCP).

✨ Features

Core Functions

Retrieve a Single Document

# Retrieve a document at the specified path
const document = await mcp.useTool("librarian", "getDocument", {
  filepath: "/daisyui/components/button.md"
});

Retrieve Multiple Documents

# Retrieve multiple documents simultaneously
const documents = await mcp.useTool("librarian", "getDocuments", {
  filepaths: ["/daisyui/components/button.md", "/daisyui/components/card.md"]
});

📚 Documentation

Directory Structure

Librarian supports the following directory structure:

• /daisyui/components: Contains DaisyUI component documentation.
• /tailwind4: Stores knowledge related to Tailwind CSS v4.
• /tutorial: Holds various tutorial content.
• /reference: Saves reference materials and API documentation.

Tag System

Core tags include:

• components: Knowledge related to components.
• styles: Knowledge related to style design.
• tutorials: Tutorial content.
• references: Reference materials.

Document Format Requirements

All documents must be in Markdown format, with metadata added to the file header, for example:

---
title: Button Component
description: Detailed description of the DaisyUI button component
tags:
  - components
  - interactive
  - button
---

Integration and Collaboration with LLM

Librarian provides the following interfaces for LLMs:

1. Tag Discovery: List all available tags to understand the knowledge classification system.
2. Document Discovery: Browse available documents to know what knowledge content is available.
3. Search: Conduct precise searches in documents according to requirements.
4. Retrieval: Obtain the full content of specific documents.
5. Context Building: Use retrieved content to provide background information for generating responses.

Error Handling

Librarian adopts the standard MCP error response mechanism, specifically including:

• INVALID_ARGUMENT: Triggered when the argument is invalid.
• NOT_FOUND: Returned when the requested document or directory does not exist.
• INTERNAL: Used when an unexpected server error occurs.

Each error response includes:

• Error code
• Descriptive information
• Optional details for debugging

Common Problem Solutions

Document Not Found

1. Check the file path: Ensure that the provided filepath or filepaths argument is correct.
2. Confirm the document exists: Verify that the target document actually exists in the specified directory.
3. Verify permissions: Ensure that the current user has access rights to the target resource.

Tag-Related Issues

1. Tag does not exist: It may be a typing error. It is recommended to check the spelling and verify the list of available tags.
2. Too many tags: Consider using more specific or broader tags for queries.

📄 License

This project follows the MIT open-source license, and the source code is hosted on GitHub:

git clone https://github.com/yourusername/librarian.git

Contribution Guidelines

Community contributions are welcome! Please refer to the CONTRIBUTING.md file to learn how to participate in the development.

Author Information

• Project Maintainer: [Your Name]
• Contact Email: your.email@example.com
• GitHub Profile: https://github.com/yourusername

Librarian is a tool still in the process of improvement. We look forward to your usage and feedback!