Overview

The Probe.dev MCP (Model Context Protocol) server enables AI assistants to directly interact with the Probe.dev API and documentation. This integration allows you to:

  • Search documentation - Find information across all Probe.dev docs
  • Analyze media files - Run FFprobe, MediaInfo, and Probe Report analysis
  • Manage usage - Check statistics and request logs
  • Authenticate requests - Automatically handle API authentication

You’ll need a Probe.dev API key to use the MCP server. Get yours at probe.dev/dashboard.

After running the initial setup with npx mint-mcp add docs.probe.dev, you’ll use npx -y mint-mcp docs.probe.dev in all your MCP client configurations.

Supported Clients

The Probe.dev MCP server works with any MCP-compatible client:

Cursor

Popular AI-powered code editor with MCP support

Claude Desktop

Official Anthropic desktop application

Continue

VS Code extension for AI coding assistance

Custom Clients

Any application supporting the MCP protocol

Installation & Setup

Step 1: Install the MCP Server

Run this command to set up the Probe.dev MCP server:

npx mint-mcp add docs.probe.dev

When prompted, you’ll see two credential prompts:

  1. Bearer Token for Probe.dev API: Enter your API token from Probe.dev dashboard
  2. API Key for Probe.dev API: Enter any placeholder value (e.g., “none” or “placeholder”)

The setup requires two prompts but only the first Bearer Token is actually used for authentication. You must enter something for the second prompt (any text will work).

You’ll be asked for your API token twice during setup - once as a “Bearer Token” and once as an “API Key”. Use the same Probe.dev API token for both prompts. After setup, you’ll use the command npx -y mint-mcp docs.probe.dev to run your MCP server locally in client configurations.

Step 2: Configure Your API Key

The MCP server needs your Probe.dev API key to make authenticated requests. Set it as an environment variable:

export PROBE_API_TOKEN="your_api_key_here"

To make it permanent, add this line to your shell profile:

echo 'export PROBE_API_TOKEN="your_api_key_here"' >> ~/.zshrc
source ~/.zshrc

Never commit your API key to version control. Always use environment variables or secure configuration files.

Client-Specific Configuration

Cursor Setup

  1. Create MCP Configuration File Create or edit the file ~/.cursor/mcp.json:

    mkdir -p ~/.cursor
    touch ~/.cursor/mcp.json
    
  2. Add Probe.dev Server Configuration Edit ~/.cursor/mcp.json with the following content:

    {
      "mcpServers": {
        "probe-dev": {
          "command": "npx",
          "args": ["-y", "mint-mcp", "docs.probe.dev"],
          "env": {
            "PROBE_API_TOKEN": "your_actual_api_key_here"
          }
        }
      }
    }
    
  3. Restart Cursor to activate the MCP server

Replace your_actual_api_key_here with your actual API key from probe.dev/dashboard. The -y flag automatically accepts prompts during server startup.

Claude Desktop Setup

  1. Locate Configuration File

    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
    • Windows: %APPDATA%\Claude\claude_desktop_config.json
  2. Edit Configuration Add the Probe.dev MCP server configuration:

    {
      "mcpServers": {
        "probe-dev": {
          "command": "npx",
          "args": ["-y", "mint-mcp", "docs.probe.dev"],
          "env": {
            "PROBE_API_TOKEN": "${PROBE_API_TOKEN}"
          }
        }
      }
    }
    
  3. Restart Claude Desktop

Continue (VS Code) Setup

  1. Install Continue Extension from the VS Code marketplace

  2. Configure MCP Server in Continue’s settings:

    {
      "mcpServers": [
        {
          "name": "probe-dev",
          "command": "npx",
          "args": ["-y", "mint-mcp", "docs.probe.dev"],
          "env": {
            "PROBE_API_TOKEN": "${PROBE_API_TOKEN}"
          }
        }
      ]
    }
    

Available Tools

Once configured, your AI assistant will have access to these tools:

Tool: search
Description: Search across all Probe.dev documentation
Example: "How do I authenticate with the API?"

Media Analysis

Tool: analyze_media_file
Description: Analyze media files using FFprobe, MediaInfo, or Probe Report
Example: "Analyze this video file: https://example.com/video.mp4"

Usage Statistics

Tool: get_usage_stats
Description: Retrieve API usage statistics and metrics
Example: "Show my current API usage this month"

Request Logs

Tool: get_request_logs
Description: Access your API request history and logs
Example: "Show my recent failed requests"

Testing Your Setup

Basic Connectivity Test

Ask your AI assistant:

"Can you search the Probe.dev documentation for authentication information?"

Expected Responses

Working correctly - The assistant should:

  • Return specific documentation excerpts for searches
  • Provide detailed media analysis results
  • Show your actual usage statistics

Not working - You might see:

  • “I don’t have access to that information”
  • “Tool not available” errors
  • Authentication failures

Troubleshooting

Usage Examples

"Find information about rate limits in the Probe.dev API"
→ Returns: Rate limit documentation with specific limits and headers

Media Analysis

"Analyze this video and tell me its codec, resolution, and duration:
https://example.com/sample.mp4"
→ Returns: Complete FFprobe analysis with video/audio stream details

Batch Operations

"Compare the codecs used in these three video files:
- https://example.com/video1.mp4
- https://example.com/video2.mkv  
- https://example.com/video3.avi"
→ Returns: Analysis of all three files with codec comparison

API Management

"Show me my API usage for the last week and identify any failed requests"
→ Returns: Usage statistics and error log analysis

Security Best Practices

API Key Security

  • Store API keys in environment variables
  • Never commit keys to version control
  • Rotate keys regularly
  • Use separate keys for different environments

Access Control

  • Limit key permissions to necessary scopes
  • Monitor usage in your dashboard
  • Set up usage alerts
  • Review access logs regularly

Advanced Configuration

Custom Tool Aliases

You can create custom aliases for frequently used operations:

{
  "mcpServers": {
    "probe-dev": {
      "command": "npx",
      "args": ["-y", "mint-mcp", "docs.probe.dev"],
      "env": {
        "PROBE_API_TOKEN": "${PROBE_API_TOKEN}"
      },
      "toolAliases": {
        "quick_analysis": "analyze_media_file",
        "docs": "search"
      }
    }
  }
}

Rate Limiting Configuration

Configure request throttling to stay within API limits:

{
  "mcpServers": {
    "probe-dev": {
      "command": "npx",
      "args": ["-y", "mint-mcp", "docs.probe.dev"],
      "env": {
        "PROBE_API_TOKEN": "${PROBE_API_TOKEN}",
        "PROBE_RATE_LIMIT": "100"
      }
    }
  }
}

Getting Help

Next Steps

Once your MCP server is working:

  1. Explore the API - Try different analysis tools and parameters
  2. Automate workflows - Use the assistant to streamline media processing
  3. Monitor usage - Set up alerts for quota limits
  4. Share feedback - Help us improve the MCP integration

The MCP server automatically stays up-to-date with the latest Probe.dev API features and documentation.