"A database that can be operated or managed" matching MCP servers:

Enables AI models to securely analyze SQLite datasets and read local text documents through a local stdio interface. It provides read-only querying, schema discovery, pagination, and timeout protections to prevent data leaks or destructive changes.

Enables SQL Server database operations (queries, table management) and file system access (read, write, directory listing) with security controls. Supports comprehensive database interactions and file management through a unified MCP interface.

Enables AI agents to send and receive files securely via JustDrop, with end-to-end encryption, QR codes, and no server-side storage.

An MCP server for reading, rendering, and searching PDF files, specifically optimized for LLMs to extract text, tables, and technical diagrams. It enables metadata retrieval, multi-format text extraction, and page-to-image rendering using PyMuPDF.

Provides secure filesystem access for AI models through the Model Context Protocol with strict path validation, file operations, directory management, and system command execution within predefined directories.

Enables agents to compress JPEG, PNG, and WebP images to a target byte size, split PDFs by page or ranges, and remove image metadata while processing local files without sending file bytes over the network.

TheWeave is a markdown-native memory architecture for Claude and any MCP-aware agent. Your agent's memory lives as plain .md files you own: a 5-verb MCP core over the vault, query-driven PageRank retrieval, bi-temporal facts (valid_from / valid_until), and a persona-as-vault model. No database and no embeddings server. The files are the memory, inspectable in your editor and versionable in git.

An HTTP MCP server that indexes large documents into exact-line-numbered sections, enabling AI models to locate, read, summarize, and edit parts of a document without ingesting the whole file.

A self-hosted MCP server that enables AI coding agents to read, edit, search, and run code in local projects with human review loops and policy controls.

Enables interaction with the WhkerDB shared PDF editor to manage collaborative rooms, file structures, and real-time annotations. It allows users to upload PDFs and images, manage note trees, and sync data across platforms through natural language.

Provides a secure workspace for AI-driven file management and Python script execution designed to run on Vercel. It enables tools for full file operations and code execution within a sandboxed environment.

Enables AI agents to inspect and work with PDF files entirely on the local machine, covering page counts, metadata, layout, encryption permissions, scan-vs-text detection, and page rendering. Generated results land in a temporary workspace as short artifact ids that can be chained between tools and exported to real folders on request.

Connect your database to your AI agent, read-only, without sharing your password. Maplith Bridge serves a folder of data files over MCP behind a token you create, and fills it from PostgreSQL under a role that can only read.

Enables local .docx files to be managed with verified writes, returning before/after evidence for every mutation, plus PDF export and lock/sync status checks.

Enforces mandatory human approval before LLM agents can edit files, create files, or execute shell commands, using session tokens and audit logging.

A locally-run server that lets an AI assistant hide files inside photos or videos, reveal concealed content, perform forensic analysis of images and videos, and convert media between formats. It reads input and writes output directly on the user's own machine, so files never need to be handed off through a separate client.

Enables AI agents to list, read and download the files attached to Gmail messages, extracting text from PDFs, Word, Excel, PowerPoint and plain-text files straight into context or saving them to disk. It runs locally with read-only Gmail access, so attachments can be summarised, checked or analysed without leaving the chat.

MCP Bridge is a local-first Linux/Windows desktop app connecting ChatGPT to approved files and Chrome through authorized OpenAI tunnels, with optional shell tools. The Glama-hosted v0.1.0 release is a restricted READ-ONLY documentation demo: five tools can access only bundled public examples, not your computer, browser, credentials, shell, or files.

Read-only MCP server providing AI access to verifiable web, GitHub, and local sources, plus a managed fantasy entity catalog, with strong security and provenance tracking.

Adds a local-first memory MCP server: long-term memory is stored as plain Markdown/JSONL files in a user-chosen directory (MEMORY_DIR) and can be shared by multiple MCP clients (Cherry Studio, Claude Desktop, DeepSeek Harness). * No database, no vector store, no network calls * Four tools: memory_read / memory_update / memory_append / memory_search * Shrink-protection + backup before overwrite *

mcp-node-zig is an MCP server for remote machines. Drop it on a build box, a homelab server, a Windows host or a VPS, and any MCP client can run commands, drive long-running processes and read and write files there over one authenticated HTTP endpoint.

An MCP server that exposes filesystem operations — listing directories, reading, writing, and searching files — as tools an LLM can discover and invoke at runtime.

Connects Claude Code, Codex or any MCP agent to a live Galley Typst project. The agent reads the document, the file tree and real Typst compiler diagnostics, then proposes edits that a person accepts or rejects in Galley; it never writes directly. No model or API key inside: it bridges the agent you already run.

Enables AI assistants to scan local disks for developer caches, large unused videos, and old logs, present candidates for approval, and move only approved items to the operating system trash so they can be restored while protecting system folders.

locidx

Pull a Figma frame once, then let Claude Code, Codex or any MCP agent read it offline with no rate limits. Six read-only tools serve the normalized node tree, design tokens and locally built SVG icons from a vault you can commit, so teammates need no Figma token.

Diagnoses why a FiveM server or resource is broken by scanning local configs, manifests, and logs, returning ranked findings with evidence. It can also propose and apply low-risk text fixes when enabled.

Enables an agent to read the user's local Messages.app database on macOS, listing chats and threads and searching iMessage/SMS history entirely on localhost, with optional AppleScript sending gated behind an explicit flag. Access is read-only and scoped-shut by default, and replies can be optionally sent when send mode is enabled.

Serves a sandboxed terminal bot's tools over stdio so any MCP client can feed, play with, rest and treat the bot, read and write files inside a restricted directory, list directories, and review or undo previous writes.

Enables an agent to create, read, and list the Markdown cards in a user-owned card book, capturing people, events, places, things, and in-jokes as durable canon. Cards are stored as plain files with YAML frontmatter that the user can inspect and edit, while keyword-triggered surfacing is handled separately by a gateway or hook.