YouTube MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HOST | No | Host for HTTP transport. Default is '127.0.0.1'. | 127.0.0.1 |
| PORT | No | Port for HTTP transport. Default is '3000'. | 3000 |
| TRANSPORT | No | Transport mode: 'stdio' for local, 'http' for Streamable HTTP. Default is 'stdio'. | stdio |
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 |
|---|---|
| get_video_infoA | Get YouTube video metadata: title, channel, description, duration (text and seconds), views, likes, comment count, upload date, and tags. Unknown values are null. Served by the provider selected with YT_PROVIDER (see get_server_info). Input: YouTube URL or video ID. Errors carry a code (e.g. ACCESS_BLOCKED, VIDEO_UNAVAILABLE, TIMEOUT) in _meta["youtube-mcp/error"] and as a [CODE] text prefix. |
| get_video_commentsA | Get YouTube video comments with replies. Returns comment text, author, likes, and date. Supports sorting by 'relevance' (top comments) or 'time' (newest). Returns error if comments are disabled. Served by the provider selected with YT_PROVIDER. |
| get_video_transcriptA | Get YouTube video transcript/captions (manual or auto-generated) with timestamps. Use 'lang' to pick a language code (e.g. 'en', 'id'), or omit it to get the first manual track (else the first auto-generated one); get_transcript_languages lists the codes. Returns both timestamped and plain text versions, paginated with startSegment/nextStartSegment. available=false with reason NO_CAPTIONS or LANGUAGE_UNAVAILABLE when captions are absent; access problems (ACCESS_BLOCKED etc.) are errors. |
| search_youtubeA | Search YouTube videos by query. Returns up to 50 results with title, channel, views, duration, and URL. Supports filtering by uploadDate (today/week/month/year) and videoDuration (short/medium/long), and sorting by relevance, date, viewCount, or rating. Served by the provider selected with YT_PROVIDER; with the default innertube provider some options are approximated and reported in warnings. |
| get_transcript_languagesA | List all available caption/transcript languages for a YouTube video. Returns language codes and names for both manual and auto-generated captions. Call this first to discover available languages before fetching a transcript. |
| get_server_infoA | Report the server version, the active provider (innertube or api) for video info, comments, and search, and which backend serves each tool. Transcript tools always use youtube-transcript-plus. |
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 6 tools
Each tool targets a clearly distinct resource and action: video metadata, comments, transcript content, language discovery, search, and server info. There is no overlap in purpose, and the transcript/language split is explicitly signposted (call get_transcript_languages first).
Most tools follow a get_<resource> pattern (get_video_info, get_video_comments, get_video_transcript, get_server_info), which is highly readable. Minor deviations exist: search_youtube uses a different verb form and get_transcript_languages breaks the get_video_ prefix used by its sibling transcript tool.
Six tools is well-scoped for a read-oriented YouTube data server, with each tool earning its place and no redundant operations. It sits comfortably in the ideal 3-15 range.
The surface covers the core read workflows well: video metadata, comments, transcripts, language discovery, and search. Gaps exist around channel-level and playlist operations, but these are workable omissions for a video-centric read-only server.