Explorar skills

Use o GitHub CLI para buscar conteúdo de URLs do GitHub em vez de buscar pela web. Ative quando o usuário compartilhar um link de issue, PR, repositório ou discussão do GitHub.

Generate reusable multi-step agent workflow blueprints. Use for trigger/action orchestration, deterministic workflow definitions, and automation handoff artifacts.

视觉设计工具箱 - Canvas / Themes / Design System / Artifacts。The unified Visual Design Engine for Antigravity.

This skill should be used when the user asks to "evaluate agent performance", "build test framework", "measure agent quality", "create evaluation rubrics", "implement LLM-as-judge", "compare model outputs", "mitigate evaluation bias", or mentions multi-dimensional evaluation, agent testing, quality gates, direct scoring, pairwise comparison, position bias, evaluation pipelines, or automated qualit

Use when the user asks how to do something, wants a skill for a task, or wants to discover and install new agent capabilities.

Use when the user wants to design a synth patch from scratch or modify an existing one - lead, bass, pad, pluck, lush keys, FX. Targets stock Ableton synths (Operator, Wavetable, Analog, Drift). Examples - "make me a wide supersaw lead", "design a sub bass", "give me a lush pad like Olafur Arnalds".

Converts a workflow description into a LangGraph node/edge graph with typed state, conditional routing, and human-in-the-loop checkpoints.

Finds and fixes bugs through systematic root cause analysis, stack trace interpretation, browser DevTools automation, CI/CD pipeline debugging, performance profiling, test pollution detection, and AI-powered error analysis. Use when the user asks to debug, fix a bug, investigate an error, analyze a stack trace, find root cause of a failure, profile performance, diagnose test failures (unit/integra

Compare model candidates using weighted metrics and deterministic ranking outputs. Use for benchmark leaderboards and model promotion decisions.

抖音文案违规检测。扫描口播稿中的引流词、极限词、医疗/金融承诺、低质标记等7类违规风险，输出结构化报告+替换建议。Use when 用户说"检测违规""检查文案""看看有没有违规词""会不会被限流""发之前帮我看看"或给出任何要发布到抖音的文案要求检查。

Scaffold frontend architecture, components, routing, and UI dependencies from project config and wireframes.

Apply RICE, MoSCoW, ICE, and effort-impact frameworks to rank options and decide what to work on next. Use when the user asks to prioritize a backlog, rank features, decide what to build first, apply RICE or MoSCoW scoring, cut scope, or compare items by effort vs impact. NOT for defining success metrics or OKRs (use outcome-orientation). NOT for making a strategic bet under uncertainty (use strat