Visualizeit

Building open-source Agent Skills and developer tools for real-world AI workflows.

构建面向真实工作流的开源 Agent Skills 与开发者工具。

I build at the intersection of AI agents, developer experience, and product interfaces. I care about turning practical workflows into open-source tools that are clear, dependable, and pleasant to use.

Selected work

Illustrator

Designed images from a regular code-capable LLM—no image-generation model required.

A lightweight, browserless Agent Skill that turns natural-language direction, content, code, and supplied images into locally rendered visuals.

Agent Skill · Local rendering · Browserless · Apache-2.0

Install · Examples · 简体中文

Current focus

• Agent Skills for practical content and development workflows
• Programmatic design and local rendering
• AI-native editors, content workflows, and collaboration
• Developer tooling and frontend infrastructure

Selected open-source contributions

• Mantine — Added missing MenuSub compound-component exports for use in React Server Components.
• oRPC — Simplified the TanStack Start integration documentation with a unified ANY handler.
• Ultracite — Fixed Oxlint filename rules for TanStack route files and added regression tests.

Writing

I write in Chinese about AI agents, developer tools, open source, and product engineering.

• Xiaohongshu / 小红书

WeChat Official Account / 微信公众号

Contact

Open to open-source collaboration, AI and developer-tooling projects, and relevant work opportunities.

Email: visualizeit@foxmail.com