nateherkai/scroll-craft

An agent skill for building premium, immersive, scroll-driven websites. Works with Codex, Claude Code, and other coding agents. Also available as a Claude...

View on GitHub ↗Jump to charts ↓

nateherkai/scroll-craft has 3.5k+ stars and 500+ forks on GitHub, is written in JavaScript, and was last pushed on Sep 30, 2026. Data as of . Signed-in members get hourly updates — create a free account.

Summary Information

Updated 1 hour ago
Added to GitGenius on October 8th, 2026
Created on August 22nd, 2026
Open Issues & Pull Requests: 9 (+0)
GitHub issues: Enabled
Number of forks: 535
Total Stargazers: 3,600 (+0)
Total Subscribers: 26 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 6.0 days
Mean response time: 6.1 days
90th percentile: 11.6 days
Tracked items: 3

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 4
New in 7 days: 1
Closed in 7 days: 0
Avg open age: 22 days
Stale 30+ days: 1
Stale 90+ days: 0

Recent activity

Opened in 7 days: 1
Closed in 7 days: 0
Comments in 7 days: 0
Events in 7 days: 0

Top labels

No label distribution available yet.

Most active issues this week

Sign in to see which issues are moving.
Sign in

Detailed Description

scroll-craft is an agent skill for building premium, scroll-driven websites that work with Codex, Claude Code, and other coding agents, also available as a Claude Code plugin.

The tool addresses the problem that most AI-generated websites fall into one of two traps: either they are bland and forgettable, or they prioritize flashy scroll animations while neglecting fundamental design craft like readable typography and responsive layouts. scroll-craft treats interaction and craft as a single unified job rather than separate concerns. It provides a design workflow, references, an animation engine, and verification tools that guide agents through building immersive scroll experiences with intentional depth planes, pointer responses, scroll transitions, and art direction tailored to each project rather than relying on generic templates.

The skill includes an approved collection of ten production websites that demonstrate different page grammars, navigation models, and information architectures, showing that the same engine can produce visually distinct results. Someone adopting this tool should expect to work with a design process that requires planning depth layers, using authentic brand assets before generating imagery, choosing between photographic compositing and 3D rendering, and art-directing mobile experiences separately. It suits teams building landing pages and portfolio sites where scroll-driven interaction is central to the brand experience, and where the goal is to avoid both forgettable output and technically impressive but poorly crafted results.

The project maintains a curated collection of worked examples with documented production workflows and design references that remain stable across updates. The skill preserves existing engine functionality while adding new design standards and video compatibility fixes. Development activity shows attention to specific design problems through detailed guidance on hero-depth planning, contact anchors, and opening-midpoint-exit state transitions rather than generic feature expansion.