Skill UI

Analyzes Claude Code and Codex usage logs to surface cost drivers such as low cache hit rates, unpaid cache writes, top-tier sub-agents, reasoning share, and bloated sessions, with evidence and what-if savings to help cut AI coding spend.

Produces case briefs (fiches d'arrêt) and decision commentaries in French administrative law using the Faouzi El Ouertani 7-section method, with systematic verification of cited rulings and texts via the LibreJustice MCP and Légifrance connectors.

Explains keyword meanings, possible search intents, and content direction from supplied context. Useful for SEO planning, ambiguous terms, and audience-specific content briefs.

This skill performs automated PR code reviews to identify decay risks, design smells, and maintainability issues. It applies principles from classic engineering books like Brooks's Law and Mythical Man-Month to provide concrete Symptom → Source → Consequence → Remedy findings. It triggers on code sharing, PR checks, or feedback requests regarding architecture and refactoring.

This tool creates concise competitor summary reports based on provided names, URLs, and evidence. It organizes observations on positioning, audience, and offerings, then provides actionable implications. Ideal for SEO analysis and market research where competitor data is available. It validates input facts and avoids fabrication, ensuring results are grounded in supplied evidence.

Find products for sale across online shops via the hosted datawebot MCP server: get titles, prices, sellers, ratings and direct links, compare offers, and run Amazon or Shopify extraction Jobs.

Build and debug Defold game cores in Lua, covering project settings, collections, game objects, script lifecycle, input, messaging, factories, and Bob build/run checks.

Gates a UI screen before it ships: lints source for inert handlers, placeholder copy and removed focus rings, checks loading/empty/error state coverage, and audits the rendered page for WCAG 2.2 AA contrast, focus, reflow and target-size failures.

This skill processes deficiency reports from Goldfish reviewers, categorizes findings into contest/escalate/propose fix/decline categories, and manages interactive resolution workflows for code review feedback with automated fixes and negotiation checkpoints.