Let AI agents watch videos and answer questions about them through any OpenAI-compatible endpoint.
Video Understanding MCP is an MCP integration. Let AI agents watch videos and answer questions about them through any OpenAI-compatible endpoint.
The published npm package is @granitebps/video-understanding-mcp, version 1.0.2. Its declared MCP transport is stdio.
The package declares VIDEO_BASE_URL (required): Base URL of an OpenAI-compatible chat completions endpoint whose model accepts video, e.g. https://host/v1; VIDEO_MODEL (required): Model (or router alias) to send videos to; VIDEO_API_KEY: Endpoint API key, sent as a bearer token; VIDEO_EXTRA_HEADERS: Extra endpoint request headers as a JSON object, such as Cloudflare Access service-token headers; VIDEO_PART_FORMAT: How the video is attached: video_url, image_url, file, or input_file; VIDEO_MAX_DURATION_SECONDS: Longest analyzed range in seconds; VIDEO_MAX_UPLOAD_MB: Largest prepared video sent to the endpoint, in MB; VIDEO_ALLOWED_DIRS: Directories local files may be read from; local files are refused when unset; VIDEO_VERIFY_MEDIA: Fail calls when the endpoint evidently ignored the video.
Check the maintainer’s instructions for account access, authentication and the supported client configuration. Once connected, inspect the tools your client receives and try a small task that matches the server’s stated purpose. The registry description does not include a verified, complete tool list.
View the publisher’s registry entry · Project repository. This profile uses version 1.0.2 from the registry. Registration identifies the publisher namespace; it does not establish that this is a first-party integration for every service it connects.
Browse related ai & ml services MCP servers for tools that cover the same kind of work.
| Variable | Description | Required |
|---|---|---|
| VIDEO_BASE_URL | Base URL of an OpenAI-compatible chat completions endpoint whose model accepts video, e.g. https://host/v1 | Yes |
| VIDEO_MODEL | Model (or router alias) to send videos to | Yes |
| VIDEO_API_KEY | Endpoint API key, sent as a bearer token | Check documentation |
| VIDEO_EXTRA_HEADERS | Extra endpoint request headers as a JSON object, such as Cloudflare Access service-token headers | Check documentation |
| VIDEO_PART_FORMAT | How the video is attached: video_url, image_url, file, or input_file | Check documentation |
| VIDEO_MAX_DURATION_SECONDS | Longest analyzed range in seconds | Check documentation |
| VIDEO_MAX_UPLOAD_MB | Largest prepared video sent to the endpoint, in MB | Check documentation |
| VIDEO_ALLOWED_DIRS | Directories local files may be read from; local files are refused when unset | Check documentation |
| VIDEO_VERIFY_MEDIA | Fail calls when the endpoint evidently ignored the video | Check documentation |
The original Chromium automation reference server — simple, screenshot-driven browser control.
Build a programmable telecommunications stack for connecting telephony services with the Internet via a cloud-based utility.
MCP server for OpenAI Agents SDK documentation
Verified SaaS, AI, and LLM pricing for 490+ tools: plans, hidden costs, TCO, and alternatives.
VectorCode exposes local repository indexing and semantic code retrieval through a Python MCP server with explicit project roots.
SWI-Prolog as a "logic calculator" for LLMs — available as an MCP server and a Python library. Eliminate the black box from LLM logical reasoning.