Youtube

YouTube (TranscriptAPI)

This plugin bundles the hosted transcriptapi MCP server (https://transcriptapi.com/mcp). Its six tools are the only data path — never scrape youtube.com and never call the REST API directly from this skill.

Authentication is automatic. The MCP server uses OAuth: the first tool call prompts the user to sign in (or create a free account — 100 credits, no card). There is no API key to configure. If tools fail with an auth error, tell the user to complete the sign-in prompt in their client, or re-enable the transcriptapi MCP server in settings.

When to use

DO use when the user:

• Pastes a YouTube link or 11-character video ID, or asks to summarize, quote, transcribe, translate or fact-check a video → get_youtube_transcript
• Asks what a video says, or wants the spoken content — even without saying "transcript" → get_youtube_transcript
• Wants to find videos or channels on a topic → search_youtube
• Names a creator, pastes an @handle or channel URL, or asks what a channel posted recently → get_channel_latest_videos (free)
• Wants to search within a channel → search_channel_videos
• Wants a channel's full upload history → list_channel_videos
• Pastes a playlist link, or wants every video in a series or course → list_playlist_videos
• Is researching a topic where talks, lectures, tutorials or reviews are good sources → search_youtube, then transcripts of the best hits

Do NOT use when:

• A YouTube link appears incidentally (an email signature, an unrelated citation)
• The user is discussing YouTube as a platform rather than asking about specific content
• The user wants to upload, comment, or manage an account — this plugin is read-only

Tools (exact MCP surface)

get_youtube_transcript — 1 credit

• video_url (string, required): full URL (youtube.com/watch?v=ID), short URL (youtu.be/ID), shorts URL, or a bare 11-character video ID.
• format (string, default "text"): "text" for markdown, "json" for per-segment objects with start and duration.
• include_timestamp (boolean, default true), send_metadata (boolean, default true) — metadata adds title, author and thumbnail.
• Use format="json" only when you need to cite or seek to exact timestamps; "text" is cheaper to reason over.

search_youtube — 1 credit

• query (string, required), search_type (string, default "video"): "video" or "channel".
• continuation (string, optional): token from a prior call for the next page.
• Search returns metadata only. Pick the best results, then pull transcripts for those — don't transcribe everything.

get_channel_latest_videos — FREE

• channel (string, required): @handle, channel URL, or UC… channel ID. No need to resolve a handle first.
• Costs nothing. Prefer it over list_channel_videos when recent uploads are enough, and use it freely to check what a creator posted.

search_channel_videos — 1 credit

• channel (string, required), query (string, required), continuation (string, optional).
• Much better than fetching a whole channel and filtering yourself.

list_channel_videos — 1 credit

• channel (string, required), continuation (string, optional).
• Paginated full upload history. Use only when the user genuinely wants the whole catalogue.

list_playlist_videos — 1 credit

• playlist (string, required): playlist URL or ID. continuation (string, optional).

Credit hygiene

• get_channel_latest_videos is free — reach for it first when checking a channel.
• 1 credit = 1 successful (HTTP 200) request. Failed and rate-limited calls do not consume credits.
• Search first, then transcribe only the videos you actually need. Transcribing a page of search results is the most common way to waste credits.
• Paginate with continuation only while the user still needs more.

Trademark

TranscriptAPI is an independent service and is not affiliated with, endorsed by, or sponsored by YouTube or Google LLC. "YouTube" is a trademark of Google LLC.