UltraSplitter — Development skill for Claude Code
自适应策略的图像主体拆分SKILL Content-aware image splitting for multimodal coding agents.
How to install UltraSplitter
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open PlevanTem/UltraSplitter and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What UltraSplitter does
自适应策略的图像主体拆分SKILL Content-aware image splitting for multimodal coding agents.
Alternatives in Development
- PaperSpine — PaperSpine is a motivation-driven skill for learning from strong academic papers, building a paper’s central a 5k ★
- /prioritize — Prioritize initiatives with context-aware financial and strategic rigor 2.3k ★
- Oh My Agent — Portable multi-agent harness for .agents-based skills, workflows, and standards-aware agent teams across Antig 786 ★
README
UltraSplitter
[English](README.md) | [简体中文](README.zh-CN.md)
[](#quickstart) [](#showcases) [](https://github.com/PlevanTem/UltraSplitter/releases) [](https://github.com/PlevanTem/UltraSplitter/issues) [](#support)
Illustrated workflow · Static view
**Turn AIGC sheets into individual assets your production pipeline can actually use.**
Image models are good at presenting a family of ideas in one image: a UI icon set, weapon collection, character turnaround, building elevation sheet, furniture board, or product-variant grid. The sheet may look finished, but the assets inside it are still trapped in one bitmap. Uneven spacing, mixed scale, labels, touching silhouettes, and edge-clipped subjects make equal slicing unreliable.
Status: v0.1.0 provides a working deterministic splitter and a provider-neutral contract for generated repair. It does not call an image-generation API by itself.
Why UltraSplitter
UltraSplitter is designed for people already generating visual assets and now need clean, separate files:
- UI and product design — split icon families, interface states, illustrations, badges, and component variants.
- Game production — extract characters, turnarounds, weapons, props, costumes, inventory art, and concept-sheet elements.
- Architecture and interiors — separate elevations, façade options, material samples, furniture concepts, and presentation-board assets.
- General creative production — turn moodboards and variation sheets into named, reviewable files for downstream tools.
Instead of asking a vi
Related Skills
Viral Content Deep Analyst
Master multimodal analysis and direct-response reverse engineering skill for viral short-form video (Whitman L
ParcelRound
Splitting work across several coding agents is easy; the seam between them is what fails. A method for the sea
Gen Images Skill
Context-aware image generation skill for Claude Code — analyzes website brand identity and generates on-brand
Pivot Web Search
Resilient multi-provider web search & content extraction for Claude Code, DeepSeek Harness, and any MCP host —
Section Review
Deep, content-aware review of one section or chapter: discover its contract, find issues per dimension, advers
Asset Map
Scan, filter, and position images from the PDF outline into chapter locations for epub/docx export. Visually a
Related Agents
Multimodal Looker
Analyze media files (PDFs, images, diagrams) that require interpretation beyond raw text. Extracts specific in
Wp Normalize
Demo-folder analyzer — converts an arbitrary multi-page HTML site into the plugin's canonical delimited demo f
Omd Codex Image
Channel-aware image materializer. Reads spec blocks in HTML/MD/JSX and materializes them through Codex's nativ