youtube
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| YOUTUBE_API_KEY | Yes | YouTube Data API v3 key. Required. | |
| YOUTUBE_TIMEOUT | No | API timeout in ms. Defaults to 30000. | 30000 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| parse_youtube_urlA | Parse any YouTube URL format and extract video ID, channel ID, playlist ID, handle, and timestamp. Supports youtube.com/watch, youtu.be, /shorts/, /embed/, /playlist, /channel/, /@handle, and bare video IDs. |
| get_transcriptB | Fetch the full transcript/captions of a YouTube video with timestamps. Returns both individual segments with timing and the full concatenated text. No API key needed. |
| search_transcriptA | Search within a YouTube video's transcript for a keyword or phrase. Returns matching segments with timestamps and surrounding context. |
| extract_chaptersA | Extract chapter timestamps from a YouTube video's description. Returns structured chapter list with titles and start times. |
| get_clean_transcriptA | Fetch a YouTube video transcript with sponsor reads, intros, outros, self-promotion, and filler removed using SponsorBlock data. Ideal for summarization. |
| get_most_replayedA | Get the "most replayed" heatmap data for a YouTube video, showing which parts viewers rewatch most. Returns intensity scores and top peaks with timestamps. Requires 50K+ views. |
| search_videosB | Search YouTube for videos with full filter support: upload date, duration, sort order, and content type. |
| get_video_metadataA | Get detailed metadata for a YouTube video: title, description, channel, duration, view/like/comment counts, tags, category, thumbnails, and more. |
| get_channel_infoA | Get YouTube channel information: title, description, subscriber/video/view counts, country, and thumbnail. Accepts channel URL, @handle, or channel ID. |
| get_playlist_itemsA | Get all videos in a YouTube playlist with metadata. Returns playlist info and video list with titles, channels, and positions. |
| get_channel_videosA | Get recent videos from a YouTube channel. Supports sorting by date or view count. |
| get_trending_videosA | Get currently trending/popular YouTube videos by region and category. |
| search_within_channelB | Search for videos within a specific YouTube channel. |
| get_video_commentsA | Fetch top-level comments from a YouTube video. Returns author, text, like count, reply count, and publish date. |
| calculate_engagementA | Calculate engagement metrics for a YouTube video: like rate, comment rate, and overall engagement rate based on view count. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 15 tools
Tools cover distinct resources and actions with minimal overlap. The only potentially confusing pair is get_transcript vs get_clean_transcript, but their descriptions clearly distinguish raw transcripts from cleaned ones.
All tools follow a consistent snake_case verb_noun pattern using get_, search_, parse_, extract_, and calculate_, making the action and target predictable across the entire set.
15 tools is at the upper boundary of the well-scoped range, but each tool serves a distinct YouTube workflow including video, channel, playlist, transcript, comments, and engagement analysis. No tool feels redundant.
The set covers most read-only YouTube needs: search, metadata, channels, playlists, transcripts, comments, trends, and engagement calculations. Minor gaps exist, such as fetching comment replies, channel playlists, or transcript language options, but core workflows are supported.