YouTube MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CACHE_DB_PATH | No | SQLite database for cache + quota | ./data/cache.db |
| TRANSCRIPT_MODE | No | Comma-separated transcript modes | provided_text |
| YOUTUBE_API_KEY | Yes | YouTube Data API v3 key | |
| GOOGLE_CLIENT_ID | No | OAuth client ID (future caption support) | |
| GOOGLE_REDIRECT_URI | No | OAuth redirect URI | |
| GOOGLE_CLIENT_SECRET | No | OAuth client secret | |
| CACHE_TTL_VIDEO_HOURS | No | TTL for video detail cache | 12 |
| MAX_DAILY_QUOTA_UNITS | No | Soft daily quota budget | 9000 |
| CACHE_TTL_SEARCH_HOURS | No | TTL for search result cache | 6 |
| CACHE_TTL_CHANNEL_HOURS | No | TTL for channel/profile cache | 24 |
| PUBLIC_TRANSCRIPT_ADAPTER_ENABLED | No | Enable public transcript adapter | false |
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.healthcheckC | Server readiness, dependency status, API reachability, and schema version |
| youtube.auth.statusA | Show whether API key and OAuth are configured |
| youtube.quota.statusA | Show estimated YouTube API quota usage by endpoint and day |
| youtube.cache.statusA | Show cache size, freshness, hit rate, and stale records |
| youtube.channel.resolveA | Resolve channel URL, handle, custom URL, channel ID, or video URL to canonical channel ID |
| youtube.channel.get_profileB | Fetch channel title, description, statistics, thumbnails, and branding metadata |
| youtube.channel.get_uploadsC | Fetch recent uploads through the channel uploads playlist |
| youtube.video.get_detailsC | Fetch video metadata, statistics, duration, and thumbnails |
| youtube.video.batch_get_detailsB | Batch video lookup by IDs or URLs with cache-aware quota handling |
| youtube.video.searchC | Search YouTube by keyword with filters for region, duration, recency, and order |
| youtube.video.performance_snapshotC | Return views, engagement rate, views per day, and packaging metrics for a video |
| youtube.thumbnail.getB | Return all available thumbnail URLs and dimensions for a video |
| youtube.strategy.channel_auditB | Channel audit with upload cadence, top videos, outliers, title patterns, and thumbnail availability |
| youtube.niche.findC | Rank niche opportunities from seed topics using search demand and competition proxies |
| youtube.transcript.getC | Retrieve transcript using approved modes (provided_text supported; OAuth captions planned) |
| youtube.transcript.analyzeC | Analyze user-provided transcript for hook, structure, CTAs, and retention signals |
| youtube.packaging.analyze_titleC | Score a title by clarity, curiosity, specificity, length, and detected formula |
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 17 tools
Each tool has a clearly distinct purpose, organized by category prefixes (e.g., youtube.channel, youtube.video). No two tools overlap in functionality; an agent can easily differentiate them.
All tool names follow a consistent pattern: lowercase with dot-separated categories and underscores (e.g., youtube.channel.get_profile, youtube.video.get_details). No mixing of conventions.
17 tools cover a comprehensive range of YouTube data and analysis operations (channels, videos, transcripts, niche analysis, etc.) without being excessive. The count is well-scoped for the server's purpose.
The tool set covers core YouTube information retrieval and analysis needs: channel details, video details, transcripts, thumbnails, search, niche analysis, and channel audit. Missing CRUD operations are outside the server's analytical scope.