Loading...
Loading...
Found 2,542 Skills
Audits web UI quality across accessibility, interaction, forms, typography, navigation, layout, performance, motion, and microcopy. Use when reviewing or refining frontend UI before merge or release, or when the user asks for a UI, UX, or accessibility audit.
Use when an AI agent should run protocols or workflow tests against kairos-dev (KAIROS MCP in this repo's dev environment). Covers AI–MCP integration and workflow-test flows; MCP-only, reports/ output.
Run 7 UI integrity checks on any URL. Catches blank renders, contrast failures, undersized tap targets, horizontal overflow, broken images, text overflow, and element overlap. Returns structured findings your agent can read and fix. Use when asked to validate UI, browser check, check before shipping, UI integrity check, accessibility check.
Build PHPStan rules, collectors, and extensions that analyze PHP code for custom errors. Use when asked to create, modify, or explain PHPStan rules, collectors, or type extensions. Triggers on requests like "write a PHPStan rule to...", "create a PHPStan rule that...", "add a PHPStan rule for...", "write a collector for...", or when working on a phpstan extension package.
Automatically diagnose and fix CI failures in the current PR. Retrieves failed logs from GitHub Actions, categorizes the failure (lint, format, type-check, test), applies targeted fixes, verifies locally, and commits/pushes. Use when CI fails after push.
Expert knowledge for Modern Java (21+) development, including Virtual Threads, performance tuning, and idiomatic clean code. Use for deep Java language/logic questions.
Use to perform market backtests with PlausibleAI Backtester, including symbol discovery, strategy validation, strategy mining, and batch execution.
Use when writing or reviewing Go code to ensure idiomatic style, up-to-date language features, and best practices.
Generates production-ready React components with TypeScript, Tailwind CSS, proper accessibility, and test scaffolding. Use when asked to create a new React component.
Encodes a continuous improvement loop for goal-seeking agents: EVAL, ANALYZE, RESEARCH (hypothesis + evidence + counter-arguments), IMPROVE, RE-EVAL, DECIDE. Auto-commits improvements (+2% net, no regression >5%) and reverts failures. Works with all 4 SDK implementations. Auto-activates on "improve agent", "self-improving loop", "agent eval loop", "benchmark agents", "run improvement cycle".