TranscriptAPI: hosted YouTube transcript + video-discovery API for AI agents. Also available as an MCP server and agent skills.
This is an n8n community node for TranscriptAPI: YouTube transcripts and video discovery for n8n. Fetch transcripts as Markdown or structured JSON, search YouTube for videos and channels, and browse channel and playlist uploads, all from one node.
n8n is a fair-code licensed workflow automation platform.
Follow the installation guide for community nodes:
- In n8n, go to Settings → Community Nodes.
- Select Install, enter
n8n-nodes-transcriptapi, and confirm.
You need a TranscriptAPI API key (starts with sk_):
- Create an account at transcriptapi.com: 100 free credits, no card (one-time); paid plans from $5/mo (1,000 credits).
- Create an API key on the dashboard.
- In n8n, create a TranscriptAPI API credential and paste the key.
The credential test uses the free video-info endpoint and consumes no credits.
One node, six operations:
| Operation | What it does | Credits |
|---|---|---|
| Get Transcript | Transcript of a video (full URL, youtu.be, Shorts, or bare ID) as Markdown with metadata or structured JSON; optional preferred language | 1 |
| Search YouTube | Search videos or channels; paginated via continuation token (~20 results/page) | 1/page |
| Get Latest Videos (Channel) | The ~15 newest videos of a channel via RSS: free, no credits | Free |
| Search Channel Videos | Search within one channel (@handle, URL, or UC channel ID) | 1/page |
| List Channel Videos | Paginated channel uploads, ~100 per page | 1/page |
| List Playlist Videos | Paginated playlist videos: PL, UU, LL, FL, and OL lists | 1/page |
Search, channel, and playlist listings return a continuation_token when more results exist. Feed it into the operation's Continuation Token field to fetch the next page; the other parameters are then ignored (the token encodes them).
- Markdown with metadata: a Markdown document with a title/channel/length header followed by the transcript text. Ready to drop into LLM prompts, notes, or docs.
- Structured JSON: timestamped segments plus the metadata block, for downstream processing.
Requires n8n version 1.0 or later and Node.js 18.10+.
TranscriptAPI is an independent product and is not affiliated with or endorsed by YouTube or Google. Use of this node is subject to the TranscriptAPI terms.