Universal Source Management System MCP Server

@YUZongmin

代码仓库

基本信息

分类

数据库

许可证

MIT

传输方式

stdio

作者

YUZongmin

关于 Universal Source Management System MCP Server

A flexible system for managing various types of sources (papers, books, webpages, etc.) and integrating them with knowledge graphs.

#agent #memory #sqlite #mcp #llm #fastmcp

Universal Source Management System MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "sqlite-literature-management-fastmcp-mcp-server": {
      "command": "docker",
      "args": [
        "build",
        "-t",
        "sqlite-lit-mcp",
        "."
      ]
    }
  }
}

Universal Source Management System MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Universal Source Management System?

A lightweight FastMCP server for managing literature, notes, entity links, and reading-list resources in a local SQLite database. It is designed for researchers and users who need to organize and query their own source library without relying on an external memory graph.

How to use Universal Source Management System?

Create a SQLite database from the provided schema (create_sources_db.sql), then install the FastMCP server with the database path via fastmcp install. Alternatively, run the server as a Docker container with a persistent volume. The server exposes a set of tools and resources for reading, adding, and updating sources, notes, and entity links.

Key features of Universal Source Management System

Local SQLite-backed source management
Notes attached to sources
Entity links between sources and named concepts
Read-only database inspection tools
MCP resources for source lookup and reading lists
Normalized identifier storage with a transitional JSON cache

Use cases of Universal Source Management System

Managing a personal literature or source database
Adding and annotating research papers or web pages
Linking sources to entities such as concepts or authors
Maintaining reading lists (unread, reading) with status updates
Batch importing sources with duplicate detection

常见问题

What identifier types are supported?

Supported types: `semantic_scholar`, `doi`, `arxiv`, `openalex`, `pmid`, `isbn`, `url`.

What entity relation types are available?

Available types: `discusses`, `introduces`, `extends`, `evaluates`, `applies`, `critiques`, `supports`, `contradicts`, `refutes`.

Does Universal Source Management System integrate with MCP Memory Server?

No, it does not integrate with MCP Memory Server or an external memory graph.

How does duplicate handling work during batch import?

Exact identifier matches return a "Source already exists" message. Title-based fuzzy matches warn about potential duplicates. Successful writes return results in input order.

What are the runtime requirements?

The server requires Python with FastMCP installed. It can run locally or as a containerized stdio MCP server via Docker. The database is a local SQLite file.