VseGPT

Provides MCP servers for VseGPT to fetch images and synthesize speech via external data sources and actions.

python

GitHub Stars

python

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.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "janvarev-mcp-vsepgt-server": {
      "command": "fastmcp",
      "args": [
        "run",
        "mcp_gen_image.py"
      ],
      "env": {
        "IMG_SIZE": "1024x1024 (optional, default 1024x1024)",
        "IMG_MODEL_ID": "ID_of_image_model_on_VseGPT (optional, default img-dummy/image)",
        "VSEGPT_API_KEY": "YOUR_VSEGPT_API_KEY"
      }
    }
  }
}

You are provisioning two MCP servers for VseGPT that run locally and communicate with external services via a standardized MPI-like interface. These servers enable image generation and speech synthesis by invoking predefined Python scripts, while keeping the models’ tool field lightweight and focused. You control when and how these capabilities are exposed to the model, and you can start, stop, or swap them as needed.

How to use

To use these MCP servers, run them as local processes and connect to them through your MCP client. Each server is exposed as a stdio-based endpoint that you launch from the command line. When your MCP client requests image generation or text-to-speech, the client will route those requests to the corresponding local server.

How to install

Prerequisites you need before starting: Python and the ability to install Python packages on your system.

Install the fastmcp package (version 0.4.1 at the time of use). Run:

pip install fastmcp==0.4.1

Prepare environment variables for the image generation server. You will set these in your shell before launching the server.

export VSEGPT_API_KEY=YOUR_VSEGPT_API_KEY
export IMG_MODEL_ID=ID_of_image_model_on_VseGPT (optional, default is img-dummy/image)
export IMG_SIZE=1024x1024 (optional, default is 1024x1024)

Launch the image generation MCP server. The server files save output images to /tmp_images as you use it.

fastmcp run mcp_gen_image.py

Prepare environment variables for the text-to-speech server. You will set these in your shell before launching the server.

export VSEGPT_API_KEY=YOUR_VSEGPT_API_KEY

Launch the speech generation MCP server. The server files save output to /tmp_images when it processes audio.

fastmcp run mcp_gen_tts.py

Additional setup notes

The image generation server supports optional model and size configuration through IMG_MODEL_ID and IMG_SIZE. If you omit them, defaults are used.

The TTS server is marked as beta. You may customize its behavior directly in the server script if you need tighter integration or new voices.

Built by

VeilStrat

AI signals for GTM teams