Claude Code + Nano Banana 2 = Crazy
This video highlights the convergence of AI code assistants and video generation tools for rapid prototyping of cinematic product websites.
What this lesson covers
Dan Kieft demonstrates how to combine Claude Code with Higgsfield's Nano Banana 2 and Kling 3.0 to generate custom websites and video animations. He walks through a multi-step workflow for creating product landing pages with smooth scroll-triggered video effects.
Key takeaways from the creator
AI-extracted notes, not independently verified product claims. Timestamp links let you check each point in the original video.
- 01:06 ↗
Nano Banana 2 is used inside Higgsfield to generate specific start and end frame images of products for consistent video generation.
- 01:36 ↗
Kling 3.0 allows users to upload start and end frames alongside descriptive prompts to animate product interactions seamlessly over short clips.
- 02:46 ↗
Combining multiple 3-second video clips in editing software allows creators to extend product animations for web integration.
- 05:14 ↗
Claude Code can be run locally via terminal interfaces to build fully custom landing pages using reference images and generated video assets.
Workflow outlined in the video
- Generate start and end frame images of your product using Nano Banana 2 inside Higgsfield.
- Animate the frames using Kling 3.0 with descriptive motion and camera prompts.
- Stitch the generated video clips together in an editor like Premiere Pro or CapCut.
- Set up Anti-Gravity and Claude Code locally to build a custom landing page around your video assets.
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
135,858 views captured 23 October 2026. ReelStack recommendation score: 37/100. Formula: radar-v5. This is ReelStack’s calculation, not a YouTube rating or a measure of factual accuracy.
Show the calculation
- 65% performance: 1.32× views vs 4 other channel videos observed at a similar age. View-evidence factor 99% (views / (views + 1,000))
- 25% freshness: 0/100 with a 14-day half-life
- Momentum pending: collecting daily snapshots; its 20% weight goes to observed performance, not free points
- 10% engagement: 2/100; likes + 4× comments, smoothed with a 500-view neutral prior