33rd of 3 in Workflow automation with AIFor image, video and design, video and animation

youtube-automation-agent

Agent that researches, produces and schedules YouTube videos after your approval

Stars
5.2k
License
MIT
Last commit
Oct 2026
Last release
Oct 2026
Language
JavaScript
Stars, last 29 weeks+5,016
22 Apr 2026: 151 stars29 Apr 2026: 162 stars6 May 2026: 177 stars13 May 2026: 189 stars20 May 2026: 946 stars27 May 2026: 1,157 stars3 Jun 2026: 1,217 stars10 Jun 2026: 1,267 stars17 Jun 2026: 1,324 stars24 Jun 2026: 1,388 stars1 Jul 2026: 1,453 stars8 Jul 2026: 1,514 stars15 Jul 2026: 1,577 stars22 Jul 2026: 1,631 stars29 Jul 2026: 1,689 stars5 Aug 2026: 1,734 stars12 Aug 2026: 1,779 stars19 Aug 2026: 2,308 stars26 Aug 2026: 2,647 stars2 Sept 2026: 3,066 stars9 Sept 2026: 3,270 stars16 Sept 2026: 3,486 stars23 Sept 2026: 3,676 stars30 Sept 2026: 3,948 stars7 Oct 2026: 4,114 stars14 Oct 2026: 4,336 stars21 Oct 2026: 4,515 stars28 Oct 2026: 4,833 stars4 Nov 2026: 5,005 stars11 Nov 2026: 5,167 stars

151 on 22 Apr 20265,167 on 11 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

Node.js app that runs a YouTube channel pipeline on your own machine: topic research, scripts, narration, images, MP4 assembly, metadata, and scheduled uploads through the YouTube Data API. Every video waits for fact review, media-rights confirmation, and your approval before publishing. A dashboard on port 3456 covers review, scene repair, and analytics-based lessons that apply only after you approve them.

Who it is for: YouTube creators who self-host and want AI-produced videos with manual sign-off

Ranked among open-source alternatives to Zapier (#3 of 3).

Strengths

  • Nothing uploads without approval; simulated or placeholder media cannot be approved
  • Eight text providers or OpenRouter, five AI video providers, plus free local slideshow rendering
  • Each generation stage is checkpointed, so a crash resumes instead of restarting
  • Built-in schedule for generation, publishing, analytics, comment sync and weekly review

Weaknesses

  • Requires Google Cloud OAuth setup; Testing-mode sign-in expires every 7 days
  • No Docker or compose files; setup is git clone, npm install and Playwright Chromium
  • Narration and images need API keys; AI video clips are paid
  • Built around a single operator and channel; multi-user support is not described

What it needs

  • no GPU
  • Needs Node.js 18.17+, YouTube Data API v3 credentials, Playwright Chromium, ffmpeg-static, text provider API key
  • Models: Gemini, OpenAI, OpenRouter, Kimi, MiMo
  • port 3456

What changed

From the release notes. No flag means the note does not say so, not that the upgrade is safe.

Also in Workflow automation with AI

See all 3
Also in Workflow automation with AI
RankProjectScore
1n8nVisual workflow automation with AI agent nodes and 1500+ integrations210.3k starscustom license94 out of 100
2ActivepiecesSelf-hosted workflow automation with TypeScript integrations, alternative to Zapier25.5k starscustom license83 out of 100