frontend▌

Comprehensive evaluation methodology guide for ADK agents covering metrics, schemas, and iteration workflows. \n \n Provides eight evaluation criteria (tool trajectory, response matching, rubric-based scoring, hallucination detection, safety) with configurable thresholds and judge model options \n Includes evalset schema documentation with multi-turn conversation support, tool use trajectory specification, and session state initialization patterns \n Outlines the eval-fix loop: start small, run

Tailwind v4 and shadcn/ui setup with CSS variables, dark mode, and semantic color theming. \n \n Automates dependency installation, Vite configuration with the Tailwind plugin, and shadcn/ui initialization \n Implements a mandatory four-step CSS architecture: root CSS variables, @theme inline mapping, base styles, and dark mode via class switching \n Includes pre-built theme provider component for dark mode toggling and persistent theme storage \n Provides troubleshooting guides for common v4 mi

Design and generate Convex database schemas with proper validation, indexes, and relationships. \n \n Provides schema templates and patterns for one-to-many, many-to-many, and hierarchical relationships using ID references instead of nesting \n Includes validator reference for all Convex types ( v.string() , v.id() , v.union() , etc.) and guidance on when to use arrays versus relational structures \n Covers index strategy with single-field and compound index examples for common query patterns \n

Smart contract security review through Trail of Bits' 5-step workflow with automated scanning, visual analysis, and property documentation. \n \n Runs Slither with 70+ detectors to identify known vulnerabilities, then checks for special features like upgradeability risks, ERC conformance, and token integration patterns \n Generates three security diagrams (inheritance graph, function visibility, state variable authorization) to reveal architectural vulnerabilities that text descriptions miss \n

Expert product design spanning UX research, UI systems, prototyping, and usability validation. \n \n Covers five-stage workflow: Discover (user research), Define (information architecture), Develop (wireframes and prototypes), Test (usability studies with 5–8 participants), and Deliver (design tokens and dev handoff) \n Includes 5-day design sprint format with daily checkpoints, user journey map templates, and card sorting methods for information architecture validation \n Provides design token

Tailwind v4 with shadcn/ui using CSS variables and @theme inline pattern. \n \n Four-step architecture: define CSS variables at root, map to Tailwind utilities with @theme inline , apply base styles, automatic dark mode switching \n Prevents 8 documented errors including color mapping failures, dark mode conflicts, @apply breaking changes, and v3 migration gotchas \n Requires @tailwindcss/vite plugin (not PostCSS), empty Tailwind config in components.json, and ThemeProvider wrapper for theme tog

Comprehensive deployment guide for ADK agents across Google Cloud platforms with CI/CD, infrastructure, and troubleshooting. \n \n Covers three deployment targets (Agent Engine, Cloud Run, GKE) with a decision matrix comparing scaling, networking, session state, and cost models \n Includes quick-deploy CLI commands, scaffolded project workflows with make commands, and full CI/CD pipeline setup via GitHub Actions or Cloud Build with Workload Identity Federation \n Provides platform-specific detai

Accessible React component library built on Tailwind CSS v4 and React Aria with compound component patterns. \n \n Provides 20+ pre-built components (Button, Modal, Form, Card, TextField, etc.) using compound composition syntax ( Card.Header , Card.Content ) \n Requires Tailwind CSS v4 and uses CSS variables with oklch color space for theme customization; no provider wrapper needed in v3 \n Semantic variant system ( primary , secondary , tertiary , danger , ghost , outline ) for consistent inten

An MCP app is a standard MCP server that also serves UI resources — interactive components rendered inline in the chat surface. Build once, runs in Claude and ChatGPT and any other host that implements the apps surface.

Diagnose and fix SwiftUI rendering bottlenecks through code review, Instruments profiling, and targeted remediation. \n \n Covers view invalidation storms, unstable list identity, expensive body computations, layout thrash, and image decoding issues with concrete code examples and fixes \n Includes step-by-step Instruments profiling workflow using the SwiftUI template to identify high body-evaluation counts and CPU hotspots \n Explains identity and lifetime mechanics, lazy loading patterns, and