tfLink MCP Integration Guide

Learn how to integrate tfLink with Model Context Protocol (MCP) to enable seamless file sharing capabilities in AI assistant applications like Claude Desktop and Cherry Studio using simple stdio transport.

What is MCP (Model Context Protocol)?

Model Context Protocol (MCP) is an open standard that enables AI assistants to securely access external data sources and tools. It acts as a bridge between large language models and various services, allowing AI assistants to:

• Access real-time data: Connect to databases, APIs, and file systems
• Execute tools: Perform actions like file uploads, data processing, and more
• Maintain security: Controlled access with proper authentication
• Extend capabilities: Add specialized functionality beyond core AI features

Why tfLink MCP Integration?

AI assistants often need to share files, documents, and generated content with users. With tfLink MCP integration, AI assistants can:

Key Features

• Single Tool: Exposes upload_file_to_get_temp_share_link for seamless file uploads
• Local Stdio Transport: Runs entirely on your local machine for security
• Smart Client: Includes filename sanitization, file size limits, and optional authentication
• Simple Configuration: Environment variables with sensible defaults
• Comprehensive Testing: Full Jest test suite and ESLint configuration

Quick Setup

Installation

npx -y temp-file-share-mcp@latest

npm install
npm run build
npm run start

Configuration Output

When you run the MCP server, it automatically generates the configuration for your MCP client:

temp-file-share-mcp version 0.1.14
temp-file-share-mcp running in stdio mode
Run via: npx -y temp-file-share-mcp@latest
Connect using an MCP client that spawns this command.

Example Cherry Studio config:
{
  "mcpServers": {
    "temp-file-share-mcp": {
      "type": "stdio",
      "name": "temp-file-share-mcp",
      "description": "Upload a local file to tmpfile.link and receive a temporary download URL.",
      "command": "npx",
      "args": [
        "-y",
        "temp-file-share-mcp@latest"
      ],
      "env": {
        "TFLINK_USER_ID": "",
        "TFLINK_AUTH_TOKEN": ""
      }
    }
  }
}

Client Configuration

Cherry Studio Setup

Copy the generated configuration to your Cherry Studio MCP servers settings:

{
  "mcpServers": {
    "temp-file-share-mcp": {
      "type": "stdio",
      "name": "temp-file-share-mcp",
      "description": "Upload a local file to tmpfile.link and receive a temporary download URL.",
      "command": "npx",
      "args": ["-y", "temp-file-share-mcp@latest"],
      "env": {
        "TFLINK_USER_ID": "",
        "TFLINK_AUTH_TOKEN": ""
      }
    }
  }
}

Cherry Studio Configuration Screenshot

Here's how it looks in Cherry Studio:

tfLink MCP server configuration in Cherry Studio interface

Claude Desktop Setup

For Claude Desktop, add this to your claude_desktop_config.json file:

{
  "mcpServers": {
    "temp-file-share-mcp": {
      "name": "temp-file-share-mcp",
      "description": "Upload a local file to tmpfile.link and receive a temporary download URL.",
      "command": "npx",
      "args": ["-y", "temp-file-share-mcp@latest"],
      "env": {
        "TFLINK_USER_ID": "",
        "TFLINK_AUTH_TOKEN": ""
      }
    }
  }
}

Usage Examples

Once configured, you can use simple prompts to upload and share files:

Example Prompts

# Upload a local file
"Please use mcp tool to upload `/Users/chris/Downloads/xx.log` and get a temporary download link."

# Upload multiple files
"Upload these files using the MCP tool:
- /home/user/report.docx
- /home/user/presentation.pptx"

# Upload with absolute path
"Upload /tmp/data.json using the MCP upload tool and share the link"

Response Example

After uploading, you'll get both structured metadata and a plain-text response:

✅ File uploaded successfully!

📁 Filename: xx.log
📏 Size: 2.4 MB
🔗 Download Link: https://d.tmpfile.link/public/2025-09-20/uuid-here/xx.log
🔗 Encoded Link: https://d.tmpfile.link/public%2F2025-09-20%2Fuuid-here%2Fxx.log
⏰ Expires: September 27, 2025 (7 days)

You can share this link with anyone. The file will be automatically deleted after 7 days for privacy.

Environment Configuration

Customize the MCP server behavior with these environment variables:

Authenticated Uploads

For longer retention and better organization, configure authenticated uploads:

# Set in your shell or MCP client env block
export TFLINK_USER_ID="your-user-id"
export TFLINK_AUTH_TOKEN="your-auth-token"

# Files uploaded with authentication are stored longer
# and organized under your user account

Development

For developers who want to contribute or customize the MCP server:

# Lint the code
npm run lint

# Run tests
npm test

# Build TypeScript
npm run build

Testing Notes

The project includes comprehensive Jest tests covering:

• Configuration validation: Environment variable handling and defaults
• tfLink client: File upload logic and error handling
• MCP tool wiring: Request/response format validation

Note: Live tfLink integration tests are intentionally excluded to avoid credential leakage.

Security & Privacy

tfLink MCP integration is designed with security and privacy in mind:

• Local Processing: MCP server runs entirely on your local machine
• Stdio Transport: No network exposure, communicates via standard input/output
• Automatic Cleanup: Anonymous uploads are deleted after 7 days
• No Account Required: Works without registration for basic usage
• Secure Transmission: All uploads use HTTPS encryption
• Filename Sanitization: Automatic file name cleaning for security

Troubleshooting

Common Issues

Getting Help

Need assistance with tfLink MCP integration?

• 📦 npm Package: https://www.npmjs.com/package/temp-file-share-mcp
• 🏠 Official Homepage: https://tmpfile.link/temp-file-share-mcp
• 🐙 GitHub Repository: https://github.com/tflink-tmpfile/temp-file-share-mcp
• 📖 Documentation: Full API documentation and examples
• 💬 Support: Community support via GitHub issues