wavecraft

Visual UI testing using Playwright MCP to interact with the Wavecraft web UI. Use this skill when testing UI components, verifying visual appearance, taking screenshots, or validating UI behavior that requires browser interaction. Requires the dev server running (cargo xtask dev).

Create new processor UI components in @wavecraft/components with default scope (component, unit test, typecheck, index exports), processorId-derived IDs, and targeted validation.

Execute required visual UI validation using VS Code integrated browser (Simple Browser) with screenshot evidence recorded in test-plan.md.

Creates a Pull Request using GitHub CLI with a PR summary file. Generates PR-summary.md in the feature-specs folder to avoid terminal pasting issues.

Ensure Wavecraft UI changes use existing design tokens and reusable patterns instead of ad-hoc values, with a fast normalization checklist for spacing, color, typography, and states.

Review Wavecraft UI changes for practical accessibility quality (semantic HTML, keyboard/focus behavior, contrast, ARIA, reduced motion) and provide fix-ready findings.

Execute Wavecraft UI/UX changes through a practical end-to-end flow: discovery, proposal, implementation, verification, and handoff with minimal-scope commits.

Run the Wavecraft CI checks locally. Prefer the native `cargo xtask` commands for speed; use Docker + `act` only when validating GitHub Actions workflows or Linux-specific behavior.

Create new custom agents (.agent.md files) for VS Code Copilot. Use when the user wants to create, scaffold, or design a new custom agent with specific roles, tools, handoffs, and instructions. Guides through agent design decisions and produces a complete .agent.md file.

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends the agent's capabilities with specialized knowledge, workflows, or tool integrations.