Only AI Ads Workflow You Need In 2026
Integrating Claude dynamic prompting with element tagging in Higgsfield Cinema Studio enables filmmakers to standardize high-concept commercial production within budget constraints.
What this lesson covers
Thomas Lundström demonstrates a modular AI advertising workflow using Higgsfield Cinema Studio, Claude, and Seedance to create cinematic product commercials. He covers element creation, prompt tagging, shot list generation, and strategies for rescuing bad AI generations during edit assembly.
Key takeaways from the creator
AI-extracted notes, not independently verified product claims. Timestamp links let you check each point in the original video.
- 02:05 ↗
Creating reference sheets that combine full product shots with close-up label views improves Seedance text and detail rendering.
- 02:47 ↗
Claude can serve as a prompting director assistant to adapt base commercial prompts for customized characters and props.
- 03:24 ↗
Crop faces out of full-body character reference sheets to prevent Seedance from getting confused when generating facial details in video clips.
- 03:58 ↗
Uploading element naming screenshots to Claude enables automatic asset tagging directly in generated Cinema Studio prompts.
- 06:13 ↗
Missing specific structural details in prompts, such as hands holding an object, causes floating element artifacts that waste generation credits.
Workflow outlined in the video
- Create dedicated element reference sheets in Cinema Studio for products, props, environments, and face-cropped character models.
- Install custom shot list director skills in Claude to automate formatted, tagged prompt creation for Cinema Studio.
- Define target commercial video lengths in Claude to limit unnecessary generation cuts and cap credit spending.
- Iteratively replace failed video shots by feeding specific patch requirements back into Claude to generate targeted fill-in prompts.
Before you use this workflow
These notes describe the source video at its publication date. Model access, pricing, connectors and interfaces may have changed. Check the original source and the provider’s current documentation before installing an add-on, connecting an account or spending credits.
ReelStack has not independently tested this workflow. Preview one representative shot and check motion, continuity and output quality before applying it to a full production. No result, cost saving or model capability is guaranteed.
Explore the tools
How to interpret the numbers
5,097 views captured 27 September 2026. ReelStack recommendation score: 38/100. Formula: radar-v5. This is ReelStack’s calculation, not a YouTube rating or a measure of factual accuracy.
Show the calculation
- 65% performance: 0.97× current views vs the channel's recent median; same-age history is not yet available. View-evidence factor 84% (views / (views + 1,000))
- 25% freshness: 4/100 with a 14-day half-life
- Momentum pending: collecting daily snapshots; its 20% weight goes to observed performance, not free points
- 10% engagement: 100/100; likes + 4× comments, smoothed with a 500-view neutral prior