How do I install mcp-sqlite-manager?
It is installed with pipx: `pipx install --force mcp-sqlite-manager`. (pipx requires Python to be available on your system.)
@jonnyhoff
基本情報
An MCP server to interact with SQLite databases
ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
mcp-sqlite-manager is an MCP (Model Context Protocol) server built with FastMCP that enables interaction with SQLite databases through structured tools. It allows users to query, update, and inspect SQLite data using a set of predefined functions, primarily designed for integration with MCP clients like Cursor.
Install the package globally using pipx (pipx install --force mcp-sqlite-manager), then configure it as an MCP tool in Cursor by adding the command mcp-sqlite-manager to ~/.cursor/mcp.json. Once connected, you can call its five tools: read_query, write_query, create_table, list_tables, and describe_table.
read_query: Execute SELECT queries and return JSON resultswrite_query: Run INSERT, UPDATE, or DELETE statementscreate_table: Create new tables from SQL schema definitionslist_tables: List all tables in the connected databasedescribe_table: Show schema info via PRAGMA table_infoIt is installed with pipx: `pipx install --force mcp-sqlite-manager`. (pipx requires Python to be available on your system.)
Add the following entry to your `~/.cursor/mcp.json` file: `{"mcpServers": {"mcp-sqlite-manager": {"command": "mcp-sqlite-manager"}}}`.
It provides five tools: `read_query`, `write_query`, `create_table`, `list_tables`, and `describe_table`. Each performs a specific SQLite operation.
The project is licensed under MIT, allowing free use, modification, and distribution.
It was created by Jonathan Hoffman (GitHub: jonnyhoff).
MCP server for interacting with Neon Management API and databases
An official Qdrant Model Context Protocol (MCP) server implementation
Local MCP server for DuckDB and MotherDuck

Connect Karma.Domains to ChatGPT, Claude, Cursor, VS Code, or another MCP client — and find a domain in plain language. The same auctions, expired, backorder, and buy-now data you use in the app, plus saved filters, note
Drop-in Apache Spark replacement written in Rust, unifying batch processing, stream processing, and compute-intensive AI workloads.
コメント