mewcp-youtube
AStheTECH/mewcp-youtube
YouTube MCP Server
Instalação
npx convez-ai AStheTECH/mewcp-youtube1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
README
A tradução para o português deste README ainda está na fila.
Your YouTube channel and content, fully accessible through AI.
A Model Context Protocol (MCP) server that exposes YouTube's Data API for searching videos, managing playlists, reading comments, and interacting with channel content.
Overview
The YouTube MCP Server provides comprehensive access to YouTube via the authenticated user's account:
- Search videos, fetch channel info, playlists, and subscriptions
- Read video details, comments, and channel activity
- Create playlists, add videos, post comments, rate videos, and subscribe to channels
Perfect for:
- AI assistants that need to search or retrieve YouTube content on your behalf
- Automating playlist management and channel organization
- Building tools that integrate YouTube data into broader workflows
Tools
Channel & Account
Returns snippet, content details, and statistics for the authenticated user's YouTube channel.
Inputs:
None
Output:
{
"kind": "youtube#channelListResponse",
"items": [{
"id": "UC...",
"snippet": { "title": "My Channel", "description": "..." },
"statistics": { "viewCount": "12000", "subscriberCount": "500" }
}]
}
Returns playlists owned by the authenticated user with snippet and content details.
Inputs:
- `max_results` (int, optional) — Maximum playlists to return (max: 50, default: 25)
Output:
{
"kind": "youtube#playlistListResponse",
"pageInfo": { "totalResults": 5, "resultsPerPage": 25 },
"items": [{ "id": "PL...", "snippet": { "title": "My Playlist", "itemCount": 10 } }]
}
Returns channels the authenticated user is subscribed to, with snippet and content details.
Inputs:
- `max_results` (int, optional) — Maximum subscriptions to return (max: 50, default: 25)
Output:
{
"kind": "youtube#subscriptionListResponse",
"items": [{
"snippet": {
"title": "Channel Name",
"resourceId": { "channelId": "UC..." }
}
}]
}
Returns recent activity events on the authenticated user's channel such as uploads, likes, and subscriptions.
Inputs:
- `max_results` (int, optional) — Maximum activities to return (max: 50, default: 25)
Output:
{
"kind": "youtube#activityListResponse",
"items": [{
"snippet": { "type": "upload", "publishedAt": "2024-01-15T..." },
"contentDetails": { "upload": { "videoId": "abc123" } }
}]
}
Videos
Searches YouTube for videos matching a query, with configurable sort order and result count.
Inputs:
- `query` (string, required) — Search query text
- `max_results` (int, optional) — Maximum videos to return (max: 50, default: 10)
- `order` (string, optional) — Sort order: relevance, date, rating, title, videoCount, or viewCount (default: relevance)
Output:
{
"kind": "youtube#searchListResponse",
"pageInfo": { "totalResults": 1000000 },
"items": [{
"id": { "videoId": "dQw4w9WgXcQ" },
"snippet": { "title": "Video Title", "channelTitle": "Channel Name", "publishedAt": "..." }
}]
}
Returns full details for a specific video including snippet, content details, statistics, and status.
Inputs:
- `video_id` (string, required) — YouTube video ID (e.g., dQw4w9WgXcQ)
Output:
{
"kind": "youtube#videoListResponse",
"items": [{
"id": "dQw4w9WgXcQ",
"snippet": { "title": "...", "description": "...", "publishedAt": "..." },
"statistics": { "viewCount": "1500000", "likeCount": "50000" },
"contentDetails": { "duration": "PT3M33S" },
"status": { "privacyStatus": "public" }
}]
}
Returns the most recent videos uploaded to a channel, sorted by date.
Inputs:
- `channel_id` (string, required) — YouTube channel ID (e.g., UCxxxxxx)
- `max_results` (int, optional) — Maximum videos to return (max: 50, default: 25)
Output:
{
"kind": "youtube#searchListResponse",
"items": [{
"id": { "videoId": "abc123" },
"snippet": { "title": "Latest Video", "publishedAt": "2024-01-15T..." }
}]
}
Returns comment threads for a specific video, with optional ordering by relevance or time.
Inputs:
- `video_id` (string, required) — YouTube video ID
- `max_results` (int, optional) — Maximum comments to return (max: 100, default: 20)
- `order` (string, optional) — Comment order: relevance or time (default: relevance)
Output:
{
"kind": "youtube#commentThreadListResponse",
"items": [{
"snippet": {
"topLevelComment": {
"snippet": { "textDisplay": "Great video!", "likeCount": 42, "authorDisplayName": "..." }
},
"totalReplyCount": 3
}
}]
}
Rates a video on behalf of the authenticated user.
Inputs:
- `video_id` (string, required) — YouTube video ID
- `rating` (string, required) — Rating value: like, dislike, or none
Output:
{
"message": "Video rated as 'like' successfully"
}
Posts a top-level comment on a video on behalf of the authenticated user.
Inputs:
- `video_id` (string, required) — YouTube video ID
- `text` (string, required) — Comment text content
Output:
{
"kind": "youtube#commentThread",
"id": "comment-thread-id",
"snippet": {
"topLevelComment": { "snippet": { "textOriginal": "Great video!", "publishedAt": "..." } }
}
}
Playlists
Returns the videos inside a specific playlist with snippet and content details.
Inputs:
- `playlist_id` (string, required) — YouTube playlist ID (e.g., PLxxxxxx)
- `max_results` (int, optional) — Maximum items to return (max: 50, default: 50)
Output:
{
"kind": "youtube#playlistItemListResponse",
"pageInfo": { "totalResults": 20 },
"items": [{
"snippet": { "title": "Video Title", "position": 0, "resourceId": { "videoId": "abc123" } }
}]
}
Creates a new playlist on the authenticated user's channel with a specified title, description, and privacy setting.
Inputs:
- `title` (string, required) — Playlist title
- `description` (string, optional) — Playlist description (default: "")
- `privacy_status` (string, optional) — Privacy setting: private, public, or unlisted (default: private)
Output:
{
"kind": "youtube#playlist",
"id": "PL...",
"snippet": { "title": "My New Playlist", "description": "..." },
"status": { "privacyStatus": "private" }
}
Adds an existing video to one of the authenticated user's playlists.
Inputs:
- `playlist_id` (string, required) — Target playlist ID
- `video_id` (string, required) — Video ID to add
Output:
{
"kind": "youtube#playlistItem",
"id": "playlist-item-id",
"snippet": { "playlistId": "PL...", "resourceId": { "videoId": "abc123" }, "position": 5 }
}
Subscriptions
Subscribes the authenticated user to a specified YouTube channel.
Inputs:
- `channel_id` (string, required) — YouTube channel ID to subscribe to (e.g., UCxxxxxx)
Output:
{
"kind": "youtube#subscription",
"id": "subscription-id",
"snippet": {
"title": "Channel Name",
"resourceId": { "channelId": "UC..." }
}
}
API Parameters Reference
| Resource | Format | Example |
|---|---|---|
| Video ID | 11-character alphanumeric | dQw4w9WgXcQ |
| Channel ID | Starts with UC | UCxxxxxxxxxxxxxxxxxxxxxx |
| Playlist ID | Starts with PL | PLxxxxxxxxxxxxxxxxxxxxxx |
IDs can be found in YouTube URLs:
Video: https://youtube.com/watch?v=dQw4w9WgXcQ → dQw4w9WgXcQ
Channel: https://youtube.com/channel/UCxxxxxx → UCxxxxxx
Playlist: https://youtube.com/playlist?list=PLxxxxx → PLxxxxx
The order parameter in search_videos accepts:
relevance— Ranked by relevance to the query (default)date— Most recently published firstrating— Highest rated firstviewCount— Most viewed firsttitle— Alphabetical by titlevideoCount— For channel searches: most videos first
Used in create_playlist:
private— Only visible to you (default)public— Visible to everyoneunlisted— Visible to anyone with the link
Durations in contentDetails use ISO 8601 format:
PT3M33S → 3 minutes, 33 seconds
PT1H2M → 1 hour, 2 minutes
PT45S → 45 seconds
Troubleshooting
- Cause: OAuth token not provided in request headers or incorrect format
- Solution:
- Verify
Authorization: Bearer YOUR_TOKENandX-Mewcp-Credential-Id: CREDENTIAL-IDheaders are present - Check your YouTube OAuth credential is active in your MewCP account
- Verify
- Cause: API calls have exceeded your request limits
- Solution:
- Check credit usage in your Curious Layer dashboard
- Upgrade to a paid plan or add credits for higher limits
- Contact support for credit adjustments
- Cause: No YouTube credential linked to your account
- Solution:
- Go to Credentials in your MewCP dashboard
- Connect your Google account via OAuth (YouTube scope)
- Retry the request with the correct
X-Mewcp-Credential-Idheader
- Cause: JSON payload is invalid or missing required fields
- Solution:
- Validate JSON syntax before sending
- Ensure all required tool parameters are included
- Check parameter types match expected values (e.g.
max_resultsmust be an integer)
- Cause: Incorrect server name in the API endpoint
- Solution:
- Verify endpoint format:
{server-name}/mcp/{tool-name} - Use correct server name from documentation
- Check available servers in your Curious Layer account
- Verify endpoint format:
- Cause: Upstream YouTube Data API returned an error
- Solution:
- Check Google service status at Google Status
- Verify your OAuth credential has the required YouTube scopes (e.g.
youtube.readonly,youtube.force-ssl) - Review the error message returned in the response — common errors:
quotaExceeded,forbidden,videoNotFound
Resources
- YouTube Data API Documentation — Official API reference
- YouTube Data API Reference — Complete method reference
- FastMCP Docs — FastMCP specification
- FastMCP Credentials — FastMCP Credentials package for credential handling