AI coding agents optimize for speed over quality, skipping specs, tests, and reviews. Agent Skills is an open-source framework from Google Chrome engineer Addy Osmani that teaches AI agents the disciplined workflows of senior engineers—within the bounds of what current models can do.

Anthropic's Claude Design introduced elegant AI-powered design generation, but required cloud hosting and vendor commitment. Open Design emerged as a local-first alternative that supports multiple AI models, runs entirely on your machine, and lets you bring your own API keys. The philosophical divide isn't about better or worse—it's about hosted polish versus self-hosted control.

AI coding assistants are powerful but fragmented—each rebuilding capabilities from scratch. antigravity-awesome-skills introduces SKILL.md standardization and a cross-platform skill library that works with Claude Code, Cursor, Antigravity, and more. The v3.0.0 Governance Update signals the transition from experimental collection to production-ready infrastructure.

A curated collection for Claude Code exploded to 42,000+ stars in less than a year, becoming essential infrastructure for developers building with AI coding agents. The taxonomy of tools—from skills to hooks to orchestrators—reveals how developers actually deploy agentic workflows in production.

AI browser automation hits a hard limit when full HTML DOMs overwhelm LLM context windows. Vercel Labs' agent-browser solves this by sending compact accessibility tree snapshots instead—cutting token usage by 90% and making web automation practical for AI agents.

Claude Code's single-agent model hits limits on complex engineering tasks. oh-my-claudecode fills this gap with automatic team orchestration—architect, executor, reviewer—across 32 specialized agents. 1,785 stars in 24 hours suggest developers needed this coordination layer.