AI Skills

Open infrastructure from Zhejiang University & OpenKG for discovering, evaluating, analyzing, and routing reusable AI agent skills across 500K+ indexed repositories.

20+ battle-tested Claude Code skills turning AI into a Fortune 500 growth, marketing, copywriting, and product operations team.

Architecture-driven guardrail and methodology pack for AI coding agents that enforces baseline planning, contract verification, and proof before done.

450+ platform-specific UI/UX design rules distilling Apple HIG, Material Design 3, and WCAG 2.2 for AI coding agents.

AI agent skill suite for instant YouTube video transcripts, semantic search, channel crawling, and playlist extraction without IP blocking.

Transform any document, codebase, or book into an interactive Obsidian StudyVault with AI-driven active recall quizzes and concept mastery tracking.

Evidence-backed codebase intelligence engine and MCP server that indexes code, dependency graphs, git history, and tests to eliminate redundant agent context retrieval.

Open-source Codex multi-agent orchestrator pairing Astra architecture with DeepSeek V4.1 Flash implementation for 97%+ compute cost savings.

Open-source AI video agent framework empowering Claude Code and coding agents to programmatically script, clone, and render short-form video variants via SVML.

Animated motion component library for React and Next.js built on Framer Motion and Tailwind CSS v4, distributed via shadcn registry and AI agent skills.

Universal prompt engineering engine and Claude skill that generates high-precision, token-efficient prompts for coding agents, LLMs, and image generators.

Vector compute framework turning complex tabular and multi-modal data into retrieval vector spaces.

Production-grade toolset integration platform connecting AI agents to 100+ business applications.

Observability, session replay, and compliance tracking platform for autonomous AI agents.

Secure sandboxed cloud execution runtime designed for AI agents to run code safely.

Search engine API built specifically for LLMs and autonomous agents to retrieve real-time facts.

Universal memory layer for AI agents providing personalized long-term context retention.

Open standard connecting AI agents to external tools, data sources, and local filesystems.

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development f

Open-source autonomous agent framework optimized for local tool calling and execution.

Open-source library connecting AI agents to real web browsers for task automation.

Turns any website into LLM-ready clean markdown and structured data.