videoindexer-mcp

videoindexer-mcp

4
entertainment_and_media

The Video Indexer MCP Server is a tool designed to facilitate interaction with Video Indexer APIs by generating and retrieving prompt content based on video insights. It supports automated API interactions and integration into larger Language Models.

Video Indexer MCP Server

A Model Context Protocol (MCP) server that provides tools and resources for interacting with Video Indexer APIs.

Features

  • vi_prompt_content: Generate prompt content from video insights
  • vi_get_prompt_content: Get the generated prompt content for a video

Use Cases

  • feed LLMs with video insights
  • Automated API interactions

example config

{
  "mcpServers": {
    "videoindexer-mcp": {
      "command": "/path/to/videoindexer-mcp/bin/python",
      "args": [
        "/path/to/your/videoindexer-mcp/src/main.py",
      ],
      "env": {
        "VI_ACCOUNT_TOKEN": "<your_video_indexer_account_token>",
      },
      "transportType": "stdio"
    }
  }
}

Replace /path/to/videoindexer-mcp with the actual path to your videoindexer-mcp directory.

Installation

  1. Clone the repository
  2. Create and activate a Python virtual environment:
# Create virtual environment
python -m venv mcp-env

# Activate virtual environment
# On Windows:
mcp-env\Scripts\activate
# On Unix or MacOS:
source mcp-env/bin/activate
  1. Install dependencies:
pip install -r requirements.txt
  1. Install the package:
pip install -e .
  1. To deactivate the virtual environment when you're done:
deactivate