← All posts

Watch Me Build a $24k/Mo Stick Figure Channel With AI (Full Course)

Jul 23, 2026
StickmanClaude CodeHiggsfield

Stick-figure channels are pulling close to $24K a month, and new ones keep popping up. Past Tense started a month before this video: 14 videos, 13K+ subscribers, one video at a million views. The format is almost stupidly simple: dead-simple visuals, fast narration, and a hook that creates instant curiosity. The example below took about 20 minutes.

1. Branding

Copy the first prompt from the free doc into Claude Code (Fable 5 or Opus models). You get five channel ideas; I picked Stick and Curious.

Connect Higgsfield (50% off): + → Connectors → Manage → Add custom connector, paste the Claude MCP link from Higgsfield’s MCP & CLI section, and name it Higgsfield. Now send the channel name and Claude makes four logo variations. Ask for a (short) channel description, create the YouTube channel, and upload both.

2. Niche research

Paste the long master prompt and give it a successful channel URL. Claude runs a deep analysis and returns a file breaking down titles, scripts, thumbnails and pacing, the stuff that took them months. Then it gives 10 ideas. I chose How did ancient humans deal with pain?

3. Script

Claude asks for length. Go over 15 minutes for real uploads (1 minute for the demo). It writes in the analyzed channel’s style, and you can ask for tweaks like “more fun and engaging.”

4. Voiceover through the ElevenLabs API

Do it manually in ElevenLabs, or let Claude do it: in ElevenLabs go to Developers → API keys → Create key, then in a new Claude Code session type “Install the ElevenLabs API” and paste the key. Type approve and the voiceover is generated and saved to your folder.

5. Visuals

Lock the style (channel-matching stick figures), 16:9, 2K. Claude makes two test styles (Nano Banana vs GPT Image 2) so you can compare; I chose Nano Banana. Then set the image cadence: I always use 2.5 seconds (25 images for one minute). Approve, and Higgsfield generates them in batches. You can watch them appear live, and they download to your folder automatically.

6. The 30-second edit

In CapCut: Settings → Edit → Image duration = 2.5 s. Drag in the voiceover, then select all images and drag them in at once. Everything syncs instantly. What takes most people two hours takes about 30 seconds. Add quiet background music and you’re done.

📝 Your notes on this video

Saved on this device only

Some links in this post are affiliate links. They cost you nothing extra and help support the channel.

New to faceless YouTube?Start with my zero → monetized roadmapNiches, titles, hooks and the monetization rules, all in one place. →

Get my master prompts free

Join the Saidox Faceless Vault: free prompts, checklists and workflows from every video, plus a community of faceless creators.

Join the free Vault →