Video Quality

An MCP server that analyzes video metadata, GOP structure, quality metrics, artifacts, and transcoding results with JSON outputs.

python

4

GitHub Stars

python

Language

5 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.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "hlpsxc-video-quality-mcp": {
      "command": "python",
      "args": [
        "/path/to/video-quality-mcp/main.py"
      ]
    }
  }
}

You can run a dedicated MCP server to analyze and compare video quality across metadata, GOP structure, artifacts, and transcoding results. This server exposes a set of tools that produce structured JSON outputs, making it easy to integrate into automated pipelines and evaluation workflows.

How to use

You run the server locally and connect an MCP client to it. Start by launching the server, then configure your MCP client to point to the local process. You can access five core tools to analyze videos and compare results: analyze_video_metadata, analyze_gop_structure, compare_quality_metrics, analyze_artifacts, and summarize_transcode_comparison. Provide video file paths as inputs, and receive consistent JSON data describing metadata, GOP structure, PSNR/SSIM/VMAF scores, artifact assessments, and transcoding summaries. Use the cursor-style configuration to register the local MCP server so the client can invoke these tools directly.

How to install

pip install -r requirements.txt

Ensure you have Python 3.10 or newer and FFmpeg installed with VMAF support on your system. FFmpeg must be available in your PATH for the server to run its analysis features.

Configuration and running details

To run the server locally and expose it via the stdio interface, use the following command to start the MCP server process:

python main.py

Register the server in your MCP client configuration with the following entry. This uses stdio communication and points to the main script that implements the MCP server. The path should be the location of your video quality MCP code.

Cursor configuration example

{
  "mcpServers": {
    "video_quality": {
      "command": "python",
      "args": ["/path/to/video-quality-mcp/main.py"]
    }
  }
}

Available tools

analyze_video_metadata

Parse video file metadata and encoding parameters, returning container format, duration, file size, bitrate, video codec, profile, level, resolution, frame rate, and pixel format.

analyze_gop_structure

Analyze GOP structure and frame type distribution, providing I/P/B distribution, GOP length statistics, and keyframe timestamps.

compare_quality_metrics

Compare objective quality metrics between a reference and a distorted video, returning PSNR (Y/U/V), SSIM, and VMAF scores.

analyze_artifacts

Assess perceptual artifacts and compute proxy quality metrics, offering artifact scores and delta changes between streams when a reference is provided.

summarize_transcode_comparison

Produce a comprehensive transcoding quality assessment with verdicts, VMAF delta, bitrate savings, key issues, and optimization recommendations.