YouTube Researcher MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| YOUTUBE_API_KEY | Yes | YouTube Data API v3 key |
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 |
|---|---|
| youtube_search_nicheA | Search YouTube for videos in a niche or category. Returns video IDs and basic metadata sorted by the specified order. Use this for targeted searches when you already have video IDs from a previous search or want raw search results without metric calculations. |
| youtube_get_video_detailsA | Fetch full metadata and calculated engagement metrics for one or more video IDs. Returns title, description, tags, thumbnails, duration, view/like/comment counts, engagement rate, view velocity, and more. Batches efficiently (50 per API call). |
| youtube_get_channel_detailsA | Fetch channel metadata including subscriber count, total views, video count, and creation date. Useful for understanding the competitive landscape — whether top videos come from large or small channels. |
| youtube_analyse_nicheA | High-level niche analysis — the primary entry point for YouTube research. Searches for videos, fetches full metadata, calculates engagement metrics, and returns sorted results with aggregate statistics including median views, engagement rates, top tags, duration distribution, publishing day patterns, and channel size breakdown. Filters out Shorts (<60s) and active live streams automatically. |
| youtube_get_thumbnailsA | Download thumbnail images for video IDs to a local directory. Enables Claude vision analysis of thumbnail style, composition, text overlays, faces, and colour schemes. Tries maxresdefault first, falls back to hqdefault. No API quota cost. |
| youtube_quota_statusA | Returns estimated API quota consumption for the current server session. YouTube Data API v3 has a 10,000 unit daily limit. A typical full niche analysis costs ~103 units. |
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
The tools are mostly distinct, but youtube_search_niche and youtube_analyse_niche overlap in purpose—both search a niche, differing only in analysis depth. The description for search_niche is also confusingly worded, mentioning video IDs from a previous search, which blurs its role.
All tools use a consistent 'youtube_' prefix with snake_case, and most follow a verb_noun pattern (e.g., get_video_details, analyse_niche). However, 'quota_status' lacks a verb, deviating slightly from the otherwise coherent naming scheme.
With 6 tools, the server is well-scoped for a YouTube research purpose. Each tool covers a necessary function—search, analysis, video details, channel details, thumbnails, and quota—without excess redundancy.
The toolset covers core research workflows: search/analyze niches, retrieve video/channel metadata, download thumbnails, and monitor API usage. Minor gaps exist, such as no direct comment retrieval or transcript access, but the main research lifecycle is well-covered.