media-downloader

Watch Media

watch_media

Download a video with captions and extract timestamped key frames to produce an on-disk manifest, letting you align what was said with what was shown for analyzing tutorials, demos, talks, or screen recordings.

Instructions

Watch a video: download it with captions and extract key frames.

Returns a manifest describing a bundle on disk - the caption transcript plus scene-change key frames whose filenames carry their timestamp, so what was on screen can be lined up against what was said. Use this to analyse a tutorial, demo, talk, or screen recording rather than merely archive it.

Check status before trusting the bundle: partial means the captions or the frames are missing, and captions.fallback then names the audio-transcriber skill to recover a transcript.

Input Schema

Name	Required	Description	Default
`video_url`	Yes	URL of the video/media to watch
`max_frames`	No	Maximum key frames to extract (1-200)
`scene_threshold`	No	ffmpeg scene-change sensitivity; lower finds more frames
`download_directory`	No	Directory to save the watch bundle	.
`subtitle_languages`	No	Comma-separated yt-dlp caption language codes	en,en-orig

Output Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

First observedv4.4.7Oct 31, 2026

TDQS

A4.2/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full load and does well: it discloses the on-disk bundle layout, that frame filenames carry timestamps, and that `status` can be `partial`. It adds a recovery path via `captions.fallback`. It doesn't mention disk footprint, network/size limits, or whether directories are created, which keeps it off a 5.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the core action, then return shape, then caveat. Every sentence carries distinct information, and there is no filler or repetition of the schema.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, so return values needn't be fully restated, and the description still covers the one thing the schema likely doesn't: the meaning of `status: partial` and the fallback path. An agent has enough to call it and interpret results, with only operational details like disk usage missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so all five parameters are already documented, and the description adds no syntax or format detail beyond them. The reference to captions and `captions.fallback` is output-related rather than parameter guidance, so baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a specific verb+resource+output: 'Watch a video: download it with captions and extract key frames.' It also explicitly contrasts itself with archiving, which separates it from the download_media / media_downloader_download_video siblings.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives a clear when-to-use list ('analyse a tutorial, demo, talk, or screen recording') and an implicit exclusion ('rather than merely archive it'), which points at the plain downloader siblings. It stops short of naming the alternative tool explicitly.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server