kimi-k3-learn
An interactive learning system that turns the 47-page Kimi K3 technical report into a single offline HTML file with running algorithms, 3D visualizations, spaced repetition quizzes, and a smart highlighting QA tool.
它通过丰富的分面过滤和嵌入在组件内的适配指令,为 AI 代理提供了独特的检索层,使代理能复用零依赖组件,避免从头生成 UI。
Forever AI Components is a structured design knowledge base and retrieval layer for AI coding agents.
The components are the database. The retrieval layer is the product.
The primary consumer is an AI agent building or improving a visual interface. The registry contains 600+ curated interface components, each structured with rich metadata across many facets and made retrievable by intent. The value is in the structure: every component is labelled, related, scored, and indexed so an agent can consistently retrieve the right one for a given design task, rather than generating UI blind.
Every component is a single .html file. Every component is a URL. Built with CSS, Canvas 2D, and SVG: zero external dependencies, no build step, no framework.
Live canvas: https://forever-components.vercel.app/infinite/ Agent entry point: https://raw.githubusercontent.com/isas1/forever-ai-components/main/agents.json Full index: https://raw.githubusercontent.com/isas1/forever-ai-components/main/infinite/components.index.json
Updated continuously: see ROADMAP.md.
An agent using this registry can:
When solving a UI task, follow this order:
useCases, technique, perfTier, style, moodperfTier: cheap unless the task explicitly requires heavy animationlabels: accessible and labels: mobile-ready by defaultcomplexity when the component will be adapted600+ zero-dependency animated web components. One HTML file each. Built for AI agents.
根据分类、Topic 和编程语言匹配的相似项目。
An interactive learning system that turns the 47-page Kimi K3 technical report into a single offline HTML file with running algorithms, 3D visualizations, spaced repetition quizzes, and a smart highlighting QA tool.

Pekingman is a multimodal agent system that integrates perception, long-term memory, human-like reasoning, emotional consistency, and real-time action to create believable NPC behavior.
A CSS-only library for adding hand-drawn arrows and handwritten labels to any webpage using a single self-contained stylesheet, no JavaScript or build tools required.