YouTube MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| YOUTUBE_API_KEY | No | Optional YouTube Data API key. Required for search, video/channel stats, comments, playlists, and trending. Leave it out if you only need transcripts. |
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 | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| searchA | Search YouTube for videos, channels or playlists by keyword, with optional filters for date, region, channel and sort order. Uses 100 of the 10,000 daily YouTube API quota units per call (other tools use about 1). |
| get_videosA | Full details and stats for up to 50 videos (IDs or URLs). 1 quota unit. |
| get_channelA | Channel info and stats by channel ID (UC...) or @handle. 1 quota unit. |
| list_channel_videosA | Latest uploads for a channel (ID or @handle), newest first, with stats. ~2-3 quota units. |
| get_commentsA | Top-level comments on a video (ID or URL). 1 quota unit. |
| get_playlistB | Videos in a playlist, in order. 1 quota unit per 50. |
| trendingA | Most popular videos right now in a region. Optional category_id (28 Science & Tech, 20 Gaming, 22 People & Blogs, 24 Entertainment). 1 quota unit. |
| list_transcriptsA | List caption tracks available for a video (ID or URL). No API key or quota needed. |
| get_transcriptA | Get a video's transcript (ID or URL). No API key or quota needed. Long transcripts are split into pages of about page_words words; when the result includes next_cursor, more of the transcript remains and can be fetched by calling again with the same arguments plus cursor. timestamps=false gives clean text for summarising; true gives [m:ss] chunks for citing moments or making chapters. start/end (seconds) limit to a section of the video. |
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 9 tools
Each tool targets a clearly distinct resource or action: videos, search, channels, channel uploads, comments, playlists, trending, and transcripts. There is no meaningful overlap that would confuse an agent.
Most tools follow a predictable get_/list_ convention, and the list_ versus get_ distinction is sensible. Search and trending are standalone verbs rather than verb_noun names, but they are still clear and not jarring.
Nine tools is a well-scoped size for a YouTube MCP server. Each tool covers a major YouTube data access need without unnecessary bloat.
The surface covers the core read-only YouTube domain thoroughly: search, video details, channels, channel uploads, comments, playlists, trending content, and transcripts. There are no obvious dead ends for common discovery and analysis workflows.