This MCP server exposes Qiniu storage, intelligent multimedia, CDN, and live streaming capabilities as a Model Context Protocol service. It enables AI clients to list buckets and objects, upload and read files, generate download links, perform image processing (resize and round corners), refresh or prefetch CDN links, and manage live spaces and streams. The server is configured to run under the uv/MCP ecosystem and can be integrated with tools like the Model Context Protocol Inspector or the Cline plugin to let an AI agent issue natural language requests that translate into MCP actions.

To use it, configure the qiniu MCP server in your MCP client setup (for example via Cline or your own MCP client). Provide the necessary Qiniu credentials and bucket information through environment variables as shown in the configuration example. Once connected, you can instruct the AI to: enumerate resources, list buckets, inspect bucket contents, upload files or content for uploading, read file contents, generate or update download links, apply image processing on stored images, refresh CDN links, create and manage live spaces and streams, and retrieve push/pull addresses for live streams. If you enable the live API key, you can also interact with live streaming workflows through the same MCP interface.

Prerequisites:

• Python 3.12 or higher
• uv package manager installed (see uv installation below)

Step-by-step installation:

1. Install uv (if not already installed) and set up a virtual environment

   • macOS/Linux: uv venv source .venv/bin/activate
   • Windows: uv venv .venv\Scripts\activate

2. Install the MCP server package in editable mode uv pip install -e .

3. Configure environment variables Create a copy of the environment template and edit it with your credentials: cp .env.example .env

   Edit .env to include:

   QINIU_ACCESS_KEY=your_access_key

   QINIU_SECRET_KEY=your_secret_key

   QINIU_REGION_NAME=your_region

   QINIU_ENDPOINT_URL=endpoint_url

   QINIU_BUCKETS=bucket1,bucket2

4. Run the MCP server in directory mode (as the documentation suggests) uv --directory . run qiniu-mcp-server

5. Optional: test with the MCP Inspector npx @modelcontextprotocol/inspector uv --directory . run qiniu-mcp-server

Prerequisites and notes:

• Ensure Python and uv are installed and available in your PATH.
• The environment variables control access to Qiniu services; guard them as you would with any credentials.
• If you modify the server configuration, restart the MCP server to apply changes.

uv installation and setup (reference):

• Mac/Linux install via the provided script or package manager
• The official uv installation guides can be consulted when needed

Tips and common issues:

• If you encounter an ENOENT error in the inspector, specify the absolute path to uvx in the command field (e.g., "/usr/local/bin/uvx").
• When using live streaming, you can provide QINIU_LIVE_API_KEY in the env to enable live-related MCP actions.
• Multiple buckets can be configured by separating them with commas in QINIU_BUCKETS; keep the list to a reasonable number to avoid lengthy responses.
• If running behind a proxy or corporate network, ensure outbound access to Qiniu endpoints and that environment variables are correctly propagated to the runtime.
• For development, you can organize additional business logic under a core/storage directory as described in the repository’s guidance.