Create and publish YouTube videos from script to upload
An n8n pipeline writes a script, adds a voiceover and stock footage, assembles the video, uploads it to YouTube and updates a Google Sheet.
Done withn8nClaude Code
- What they did
- A Telegram message to a bot (or a schedule) starts the workflow. GPT-4o-mini returns a JSON script. Each voiceover segment goes to ElevenLabs, Pexels supplies footage with an image-generation fallback, and Shotstack renders the video from a JSON timeline. The video uploads to YouTube as unlisted for review, then Telegram and a Google Sheet are updated.
- How it went
- The author reports about $0.30 of voiceover and a 3-5 minute render for a 10-minute video. No view or revenue results are given, and the long-term daily render load is untested.
- Worth knowing
- YouTube's API quota allows only about six uploads a day (1,600 units each against a 10,000 limit), so iterating burns quota fast.
Try it yourself with n8n
Build an n8n workflow that takes a topic from a Google Sheet row, writes a script for a [length]-minute YouTube video about [my niche], generates a voiceover, pulls matching stock footage, assembles the video, and writes the title and description. Update the sheet with the status and file link. Stop before uploading and show me the finished video and metadata so I can approve the YouTube upload.
3 more accounts of this
Source: community.n8n.io · Undated
Five of these in your inbox every morning
The best things people got an AI agent to do, each with the prompt to try it.
More like this
Turn tickets into merged pull requests
Pulls tasks from GitHub Issues or Linear, then runs Claude Code or Codex in isolated Kubernetes pods to carry each one through to a PR.
Debug frontend UI bugs with a browser agent
A frontend developer pairs Claude in Chrome with Claude Code to open the app, click through it, read the DOM, check console errors and network requests, and fix UI bugs.
Run 24/7 WhatsApp support and track application status
A consulting firm used n8n multi-agent workflows to answer WhatsApp questions, collect information, parse uploaded documents and notify clients of status.
Produce a track in Ableton Live by talking
An MCP server that lets Codex drive Ableton Live, used to write and mix a full song from chat messages.
Curate a weekly newsletter from RSS feeds
OpenClaw monitors RSS feeds, scores items, compiles a digest with intros and sends a Telegram preview; the owner approves and it sends.
Generate illustrations in a learned visual style
A designer Hermes profile learns a style from reference designs, writes a skill for it, then crafts prompts and generates and saves images via Nano Banana.