"Using a cursor to code or implement functionality" matching MCP servers:

Enables saving files and folders to a local NAS over NFS with automatic share discovery, smart mounting, and seamless file transfers.

A local agent that monitors your system clipboard, automatically uploads copied images to Supabase Storage, and replaces the clipboard content with the public URL for immediate sharing.

Connects any MCP client to a Nextcloud instance through official WebDAV and OCS APIs, enabling browsing, searching, uploading, downloading, copying, moving, and deleting files and folders, plus managing favorites, public links and user/group shares, and the trash bin. Authentication uses the browser-based Login Flow with a saved app password.

An MCP server that gives an AI agent controlled access to a Selectel cloud account: OpenStack cloud servers, S3 object storage, and account billing.

Enables editing existing Docs, Sheets, and Slides in place, plus commenting, sharing, moving, and exporting files with native Google URLs returned. Provides 29 tools across Drive, Docs, Sheets, and Slides, using OAuth keys kept locally in ~/.gdrive-mcp/.

Enables agents to browse, read, write, upload, download, move, rename, trash and restore files in iCloud Drive by talking to iCloud directly, so it works whether or not files are synced locally. Reads prefer a hydrated local copy when one exists, while every write goes straight to iCloud so all devices see changes at once.

Lets a coding agent upload files from disk to a Stickafile portal and return a shareable link, with the bytes going straight from disk to storage without entering the model's context. Also lists the portals a token can push to.

Enables AI clients to search a cloud-drive resource search engine and turn selected results into saveable share links, using a two-step search-then-retrieve workflow.

Enables AI agents and MCP-compatible clients to search, inspect metadata for, and download files from a personal Microsoft OneDrive account via Microsoft Graph, using only read-only HTTP GET requests.

Uploads files, text, or base64 binary content to Aliyun OSS as objects with MD5-based keys, returning status, size, key, and URL.

Enables MCP-compatible AI assistants to manage iDrive backup/restore operations, including listing devices, browsing and downloading backed-up files, and performing mutations like creating folders, deleting files, or restoring from trash.

Enables creating storage buckets and uploading media files to Supabase Storage through natural language, with no delete or overwrite capabilities.

Enables Claude Code to manage Gmail, Google Calendar, and Google Drive via a dedicated Google account, keeping personal data separate.

Enables MCP clients to upload local or remote images to PicGo-configured image hosts, reusing existing PicGo configs without the desktop app. Supports batch uploads, custom naming, and status retrieval.

Enables users to upload files and generate tracked, shareable links directly from AI agents like Claude Desktop or Cursor. It supports publishing various file formats including text, PDFs, and images, while providing tools for artifact management and analytics.

Enables agents to list, read, write, create, move, delete files and folders, create public share links, and search filenames in a single Nextcloud account via WebDAV and OCS, using app-password credentials from environment variables.

Enables MCP clients to upload local files, create default 24-hour controlled download links, and list, inspect, or permanently delete owned files and shares via a dedicated API key.

Enables AI agents to publish HTML pages, files, or directories as public permanent links, with optional token-based updates to keep the same URL.

Enables AI assistants to interact with Seafile cloud storage for managing self-hosted files and directories. It supports operations such as reading, writing, moving, and searching across libraries using either account-based or library-specific authentication.

Enables agents to record their Linux X11 display while doing browser or computer-use work, then return a replay link when the task finishes without sending the video itself to the model.