# AI Video Check [Health: Active]

**Category:** 🎥 Multimedia Process  
**Repository:** https://github.com/aivideocheck/aivideocheck-mcp  
**GitHub Stars:** 0  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/ai-video-check

## Description
Check whether a video was generated by AI (Sora, Veo, Kling, HeyGen and others).

## Tools
Capabilities this server exposes over MCP:

- **check_video** — Check whether the video at a public link (TikTok, Instagram, X, YouTube, Facebook
    or a direct file URL) was generated by AI. Spends credits: one per started minute of
    video. Waits up to about a minute; if the check is not done by then, returns its id
    for get_check.
- **get_check** — Get the status or result of an earlier check by its id. Free. Results are kept
    for 24 hours.
- **get_balance** — Show how many AI Video Check credits the account has left and the limits of its key.
    Free.

## Claude Desktop Quick Installation
No confirmed install details yet, so no config is published. Follow the setup steps in the repository above.

## Documentation

## What AI Video Check does

The AI Video Check MCP server gives an MCP client access to AI Video Check, a service that evaluates whether a video may have been generated by AI systems such as Sora, Veo, Kling, MiniMax, or HeyGen. Results include a probability for AI generation and separate scores for the visual and audio portions of the video.

The server belongs in the 🎥 Multimedia Process category and is useful for media review workflows, content triage, and agent-assisted investigation. It does not establish that a video is authentic or synthetic. A low score means the service found no signs of generation; it does not prove that the footage is genuine. The detector also does not identify editing, visual effects, or face swaps applied to real footage.

## How it works

The AI Video Check MCP server supports two connection patterns. A remote Streamable HTTP endpoint is available at `https://api.aivideocheck.net/mcp`; clients send the API key in an `Authorization` header. A local Python process can instead run from a cloned repository and use an API key supplied through its environment.

For online media, `check_video` accepts a public TikTok, Instagram, X, YouTube, or Facebook link, as well as a direct file URL. It can wait for a result for roughly one minute. If processing has not finished, the tool returns an identifier that can be passed to `get_check`. Completed results remain available for 24 hours.

In local mode, `check_file` uploads an eligible video from the computer for analysis. The local implementation limits uploads to files that appear to be video containers by name and structure, including formats such as MP4, MOV, WebM, MKV, and AVI. Uploaded files are deleted when the check ends.

## Tools and capabilities

- `check_video`: Starts an analysis of a public video URL. It costs one credit for each started minute of video.
- `check_file`: Analyzes a qualifying local video file in local-server mode and uses the same per-minute credit model.
- `get_check`: Retrieves the status or result of an earlier check using its ID. This operation is free.
- `get_balance`: Displays remaining credits and the limits associated with the API key. This operation is free.

## Setup and configuration

Create an API key through the AI Video Check account dashboard. The first key includes 20 free API credits, and anyone who obtains the key can spend its credits. Keys can be revoked from the same dashboard.

For local operation, install the repository requirements in a Python virtual environment and configure `AVC_API_KEY`. `AVC_API_BASE` can change the API address from its default, and `AVC_MCP_WAIT_S` controls how long `check_video` waits before returning a check ID. The remote endpoint receives the key from the client request instead of the server environment.

The implementation is released under the MIT license. API usage consumes the same credits as the website and is subject to the service’s terms and privacy policy.

_Full upstream README: https://allmcps.com/mcp/ai-video-check/readme_