All servers / Media / FrameCap MCP server
FrameCap MCP server
FrameCap for YouTube (getframecap.com): an MCP server and command that let any AI agent capture YouTube videos and video files (frames + transcript) with the Fr
Tested by ConnectATDScore 82 out of 100. Tested 2026-11-19 on package version 1.0.7. Package: framecap-mcp
Score breakdown
| Runs | 18 / 20 | |
|---|---|---|
| Tool descriptions | 20 / 20 | |
| Package safety | 16 / 20 | |
| Maintenance | 20 / 20 | |
| Adoption | 8 / 20 |
What we saw
- Starts and answers the MCP handshake: yes (0.7 seconds including download)
- Tools offered: 3
- Last published: 2026-11-15
- Weekly downloads (npm): 948
- License: SEE LICENSE IN LICENSE · Direct dependencies: 0
Want this server set up for your team?
ConnectATD builds AI automation and wires up MCP servers like this one. Tell us what you are trying to do.
Talk to ConnectATDTools and how they are described
Agents choose tools from these descriptions. Under each name you can see how many of its parameters are explained.
capture_youtube_video13 of 13 parameters describedCaptures a YouTube video (or part of it) in the user's browser with FrameCap Pro and returns the files: contact-sheet images (each holds several frames in time order, with a t=MM:SS chip on every frame), transcript.txt (timestamped) and README.txt (how to read them). Use it to see what a video shows on screen, e.g. code, commands, slides or UI steps from a tutorial. To read small text, crop one frame out of a sheet at full size. Answers within about 40 seconds: if it's still capturing, the reply gives a capture_id; call get_capture with it until it's done.
capture_local_video14 of 14 parameters describedCaptures a video file on the user's computer (MP4, WebM, MOV, M4V, MKV or AVI) with FrameCap Pro: the same files as capture_youtube_video, with the spoken words from FrameCap's speech-to-text (60 languages) or from a subtitle file. The file stays on the computer. The first time, FrameCap downloads its speech model once (a few hundred MB), and speech-to-text on a long file takes minutes. Answers within about 40 seconds: if it's still capturing, the reply gives a capture_id; call get_capture with it until it's done.
get_capture1 of 1 parameter describedWaits up to about 40 seconds for a capture that capture_youtube_video or capture_local_video started, then returns its files, or says it's still working and what it's doing. Call it again until it's done.
Tested in a sandbox with an empty environment: no API keys, a throwaway folder, nothing from a real account. A passing test shows the server runs and responds. It is not a security audit.