MCP Personal

MCP Personal is an MCP server collection that provides local productivity tools and system utilities to AI assistants. It connects model context protocol clients, such as Claude Desktop or Claude Code, to several focused utilities: a file name search engine using fd and fzf, a fuzzy content and document search tool powered by ripgrep and fzf, a configurable SQLite database interface, and a Python port of the sequential thinking reasoning server. Developers, researchers, and power users employ these tools to inspect local codebases, extract text from PDFs, query SQLite databases safely, and manage multi-step structured reasoning workflows directly from their AI chat interfaces. By running scripts locally, MCP Personal enables AI models to interact with existing operating system binaries and local datasets. The suite includes optional document extraction features via ripgrep-all and PyMuPDF, as well as standalone CLI modes for testing each module independently. Its read-only default policies on database tasks provide security against unintended file modifications during automated workflows.

Category: Notes & Knowledge Management

Tags: automation, personal-management, productivity, utilities

Visit MCP Personal

How to install and configure MCP Personal

1. Ensure Python 3.10 or higher is installed, along with uv or pip, and install required system tools such as fd, ripgrep, and fzf. 2. Clone the repository and navigate into the folder: bash git clone https://github.com/dashed/mcp-personal.git cd mcp-personal chmod +x *.py 3. Install dependencies using uv: bash uv sync --dev 4. For Claude Code CLI, add individual servers using the user scope: bash claude mcp add file-search -s user -- $(pwd)/mcp_fd_server.py claude mcp add fuzzy-search -s user -- $(pwd)/mcp_fuzzy_search.py claude mcp add sqlite -s user -- $(pwd)/mcp_sqlite_server.py claude mcp add sequential-thinking -s user -- $(pwd)/mcp_sequential_thinking.py 5. For Claude Desktop, add the desired server paths under mcpServers in ~/Library/Application Support/Claude/claude_desktop_config.json.

What you can do with MCP Personal

• Locating project files quickly by partial name matches across large directories using fd and fzf. - Searching text patterns across local files, PDF documents, or office files with ripgrep and fzf filtering. - Querying local or in-memory SQLite databases in read-only mode to extract schema definitions and execute SQL commands. - Executing multi-step reasoning, hypothesis generation, and thought revision using the integrated sequential thinking protocol. - Extracting page counts, outlines, and text content from PDF files directly via PyMuPDF tools.

Key facts

• https://github.com/dashed/mcp-personal
• Notes & Knowledge Management, Other & General Purpose, Project & Task Management
• automation, personal-management, productivity, utilities

Part of MCP Servers

What is MCP Personal?

MCP Personal is a set of Python-based Model Context Protocol servers designed for personal productivity. It bundles utilities for fuzzy file name searching, content search across code and documents, SQLite database queries, and structured sequential reasoning into modular scripts that connect to compatible AI clients.

What external dependencies does MCP Personal require?

The project requires Python 3.10 or higher. For file searching, fd and fzf must be installed. The fuzzy search server relies on ripgrep and fzf. If you want optional PDF searching and page extraction, you can additionally install ripgrep-all, pandoc, and PyMuPDF.

Which MCP clients work with MCP Personal?

MCP Personal works with any standard Model Context Protocol client. The documentation explicitly details integration workflows and configuration steps for Claude Code CLI and Claude Desktop, and servers can also be inspected with the MCP Inspector tool.

Can the SQLite server execute write operations?

By default, the SQLite server operates in read-only mode to prevent unintentional modifications. Write operations can be explicitly enabled at startup by appending the --allow-writes flag to the server command or by setting the MCP_SQLITE_ALLOW_WRITES environment variable to true.