- Home
- MCP servers
- yunpan
yunpan
- typescript
4
GitHub Stars
typescript
Language
6 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
You connect to 360 AI 云盘 MCP to let AI models manage your 云盘 files through a standardized MCP protocol. This enables intelligent, automated file operations such as listing, uploading, downloading, searching, and sharing directly from AI-powered workflows.
How to use
Use an MCP client to connect to the 360 AI 云盘 MCP service and perform file operations end-to-end. You can choose between a local stdio setup (running a small MCP server locally via a command) or remote access over HTTP/SSE endpoints. Once connected, you can ask the AI model to list directories, upload files, download links, save content via URLs or text, create folders, rename or move items, generate share links, and fetch your user information. Authentication is done with an API key that you provide when configuring the MCP client. Use these capabilities to enable intelligent file management, backups, and content-aware workflows within your AI applications.
How to install
# Prerequisites
- Node.js 18+ (for stdio MCP server) or a runtime that can reach HTTP/MCP endpoints
- Access to the internet from the machine running the MCP client
# Option A: Use a local stdio MCP server (via npx)
# This runs a local MCP server that your MCP client can connect to using stdio
npx -y @aicloud360/mcp-server-disk
# If you need to set an API key, export it for the process
export API_KEY=YOUR_API_KEY
# Alternative start (if the snippet requires different args), you can adapt as shown in the example
# Option B: Connect via HTTP (remote MCP server)
# Your MCP client points to a remote MCP URL for the 云盘 service
# Example: configure the client to use the provided HTTP endpoint
{
"mcpServers": {
"mcp-server-disk-http": {
"url": "https://mcp.yunpan.com/mcp?api_key=YOUR_API_KEY"
}
}
}
# Option C: Connect via SSE (remote MCP server with server-sent events)
# Use a streaming endpoint for real-time updates
{
"mcpServers": {
"mcp-server-disk-sse": {
"url": "https://mcp.yunpan.com/sse?api_key=YOUR_API_KEY"
}
}
}
Additional sections
Configuration and access details are provided below. Make sure to replace placeholders like YOUR_API_KEY with your actual API credential. You can choose from three connection styles: stdio for a local server, HTTP for a remote API, or SSE for real-time streaming updates.
You can also set up a separate environment variable for the API key if you prefer not to export it in the shell. The key format starts with yunpan_ as the prefix, for example yunpan_yourkey.
Available tools
file-upload-stdio
Upload local files to the 云盘 at a specified path. Supports multiple files and a default root upload path.
file-download-stdio
Obtain a download link for a 云盘 file and optionally download it to a local directory. Can fetch either just the link or perform the download.
file-list
List files and folders within a 云盘路径. Supports pagination.
file-search
Search 云盘 contents by keyword and optional category. Returns matching results with pagination.
file-save
Save a file to 云盘 from a URL or from provided text content, placing it under a specified upload path.
video-download
Download one or more videos from URLs into 云盘 with real-time progress monitoring and batch support.
make-dir
Create a new directory within 云盘 at a given path.
file-share
Generate share links for specified 云盘 files.
file-move
Move files or folders within 云盘 to a new location.
file-rename
Rename a file or folder within 云盘.
user-info
Retrieve user personal information from 云盘.