Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.
Guides agents to validate visible Pi UI changes with the pi-ui-render CLI. Use when creating or editing Pi extensions, themes, renderers, monkeypatches, or TUI components that affect tool rows, user/assistant messages, custom messages, widgets, footers, editors, dialogs, status, or working indicators.
Reviews UI copy and microcopy, then creates a self-contained DocuSketch°-style HTML report with specific rewrites. Use when the user asks to review, audit, or critique interface text, button labels, errors, empty states, onboarding copy, tooltips, modals, forms, pasted strings, screenshots, URLs, Figma/spec links, or code with user-facing strings.
Creates new agent skills and updates existing skills so they follow the documented SKILL.md format and Addy Osmani-style skill guidance. Use when the user wants to create a skill, change a skill's behavior, audit a skill, improve a skill's trigger description, convert guidance into a skill, or make an existing skill more workflow-driven and verifiable.
Runs a full design critique across UI design, interaction design, and content design, then creates a self-contained DocuSketch°-style HTML report. Use when the user asks for an end-to-end UI review, design critique, screen audit, component critique, or shares a screenshot, URL, file path, code snippet, Figma/spec link, or flow and wants broad design feedback rather than one lens only.
Reviews how an interface behaves, communicates state, and supports user flow, then creates a self-contained DocuSketch°-style HTML report. Use when the user asks to audit UX, navigation, wayfinding, feedback, affordance, flows, states, mental models, usability, or whether a screen or prototype makes sense to users.
Guides agents through safe targeted JJ commit preparation, execution, and verification. Use when preparing, reviewing, or making a Jujutsu commit with jj.
Reviews UI visual design and composition, then creates a self-contained DocuSketch°-style HTML report. Use when the user asks to critique or improve how a screen, component, screenshot, mockup, code path, URL, or design comp looks: hierarchy, layout, alignment, density, typography, color, visual affordance, polish, or reduction.