n8n is the dominant self-hosted automation platform in 2026, and YouTube-transcript workflows are one of the most common community requests. This tutorial wires Scavio's transcript endpoint into an n8n HTTP Request node and adds transcripts to any workflow that touches a YouTube URL.
Prerequisites
- n8n self-hosted or cloud
- A Scavio API key
- A YouTube URL source (RSS, manual, Airtable)
Walkthrough
Step 1: Add a Scavio credential in n8n
Settings > Credentials > New > Header Auth. Name=x-api-key, value=your key.
Text
Name: Scavio API Key
Header: x-api-key
Value: sk_live_...Step 2: Add an HTTP Request node
POST to the Scavio transcript endpoint.
Text
URL: https://api.scavio.dev/api/v1/search
Method: POST
Auth: Scavio API Key (credential)
Body: { "platform": "youtube_transcript", "query": "{{ $json.video_url }}" }Step 3: Parse the transcript
n8n's Set node extracts the transcript array.
JavaScript
// Set node expression
{{ $json.transcript.map(seg => seg.text).join(' ') }}Step 4: Chain to an LLM node
Feed transcript to OpenAI or Anthropic node for summary.
Text
Prompt: Summarize this YouTube transcript in 5 bullets:
{{ $json.transcript_text }}Step 5: Write back to destination
Airtable, Notion, or Slack.
Text
Airtable node > Update row > Fields: summary = {{ $json.summary }}Python Example
Python
import os, requests
API_KEY = os.environ['SCAVIO_API_KEY']
def transcript(video_url):
r = requests.post('https://api.scavio.dev/api/v1/search',
headers={'x-api-key': API_KEY},
json={'platform': 'youtube_transcript', 'query': video_url})
segs = r.json().get('transcript', [])
return ' '.join(s['text'] for s in segs)
print(transcript('https://youtube.com/watch?v=dQw4w9WgXcQ')[:500])JavaScript Example
JavaScript
const API_KEY = process.env.SCAVIO_API_KEY;
export async function transcript(videoUrl) {
const r = await fetch('https://api.scavio.dev/api/v1/search', {
method: 'POST',
headers: { 'x-api-key': API_KEY, 'Content-Type': 'application/json' },
body: JSON.stringify({ platform: 'youtube_transcript', query: videoUrl })
});
const { transcript } = await r.json();
return transcript.map(s => s.text).join(' ');
}Expected Output
JSON
Full transcript text in n8n $json. Typical video: 30-second extraction. Cost per transcript: ~5 credits.