Dev Skills

GSAP animation for React, Next.js and vanilla JS. Use for scroll-driven animation, pinning, parallax, timelines, SplitText or SVG animation, Flip transitions, or useGSAP with SSR-safe cleanup.

Stage + commit changes in semantic + reviewablegroups. Use when user wants to commit changes, organize commits, or clean up a branch before pushing.

Dash + DMC v2—MantineProvider, styles, Styles API, callbacks (pattern, clientside, bg), Pages, charts, modals, full DMC. Dashboards, forms, viz.

Motion/CSS animation for React, Vue, Base UI, and vanilla JS via the Motion AI Kit: transitions, springs, easing curves, CSS spring generation, Motion API docs search, and MotionScore performance audits. Use to implement web animation or to grade and fix animation jank. Covers motion mcp, MotionScore, spring bounce, easing curve, linear() easing, transition preview, jank audit, will-change, and compositor cost.

Manage shadcn/ui components: add, search, fix, style and compose, with registry and component docs. Applies to any project with a components.json. Triggers on "shadcn init" or a --preset code.

Zod v4 TypeScript schema design, Zod 3 to 4 migration, validation boundaries, errors, codecs, JSON Schema/OpenAPI, metadata, Zod Mini/Core, RHF/tRPC/Hono/Next integrations, and rule-aware audits for Zod 4.4.3.

Bun development/runtime: adoption, Node migration, lockfiles/package-manager drift, bunfig, test, build, and command policy.

Inspect package/upstream source with opensrc: internals, exact paths, version diffs, and upgrade audits for npm, PyPI, crates, and Git; not general web.

Bounded subagent delegation and synthesis. Use for explicit subagent, delegation, fan-out, spawn policy, custom agent routing, wait behavior, and evidence-first synthesis.

Expert guidance for building chat UIs with AI SDK React hooks. Use when: (1) Building chatbots with useChat hook, (2) Implementing tool UIs with server/client execution, (3) Handling message persistence and stream resumption, (4) Creating generative UI with React components, (5) Integrating with Next.js/Node/Fastify/Nest backends, (6) Using useObject for streaming structured data. Triggers: useChat, useObject, useCompletion, chat UI, chatbot, generative UI, message persistence, @ai-sdk/react, UIMessage, tool parts, streaming UI, toUIMessageStreamResponse.

Design engineering principles for making interfaces feel polished. Use when building UI components, reviewing frontend code, implementing animations, hover states, shadows, borders, micro-interactions, enter/exit animations, choosing or reviewing icons, or any visual detail work. Triggers on UI polish, design details, make it feel better, feels off, stagger animations, border radius, optical alignment, image outlines, box shadows, icons, icon stroke weight, icon states, motion restraint.

Firecrawl CLI for search, scrape/crawl, map, extract, interact, monitor, and parse local PDF/DOCX/XLSX/HTML.

Validation routing and test-failure triage for any repository. Use to select the right gates, author deterministic tests, reproduce CI failures, or diagnose flaky suites.

Run a Codex-only structured code review closeout for local, branch, or commit diffs. Use when the user asks for autoreview, Codex review, structured closeout review, final review before commit/ship, or review after non-trivial code edits.

Audit Dash apps for callback hazards, state flow risks, layout and accessibility issues, and Dash-specific UX regressions. Use when the user asks for a Dash callback review, Dash UI audit, Dash web-interface audit, or a read-first remediation plan for a Dash app. Do not use for generic React or Next.js UI review.

Align docs with code. Detect drift and update ADRs, specs, README, AGENTS. Use after implementation or for docs drift reviews. Not for code review fixes or platform architecture.

Read-only design-system drift audit: compare an implementation against its own documented design contract, prove each deviation, and plan verified refinements without replacing product identity. Scope is system consistency only -- it deliberately excludes accessibility, semantics, and copy, so it is not a general UI review. Use when asking whether a surface has drifted from the design system. For a holistic cross-discipline interface review use better-interface.

Sentry CLI atoms for safe issues, traces, releases, projects, sourcemaps, and deliberate mutations.

Expert guidance for Vercel's Streamdown library - a streaming-optimized react-markdown replacement for AI applications. Use when: (1) Rendering AI-generated markdown from useChat/streamText, (2) Building chat UIs with streaming responses, (3) Migrating from react-markdown to streaming-friendly rendering, (4) Configuring code blocks (Shiki), math (KaTeX), diagrams (Mermaid), (5) Handling incomplete markdown during AI streaming (remend preprocessor), (6) Customizing markdown styling with Tailwind/CSS variables, (7) Securing AI output with rehype-harden (link/image protocols). Triggers: Streamdown, streaming markdown, AI chat markdown, react-markdown replacement, AI Elements Response, incomplete markdown, remend, Shiki themes, Mermaid diagrams, KaTeX math, rehype-harden, isAnimating, markdown streaming.

Generic Vitest/TypeScript testing: write or repair tests, triage flakes, tune monorepo config, and reduce local/CI test time.

Expert guidance for AI SDK Core: text generation, structured data, tool calling (tool/dynamicTool), MCP integration (createMCPClient, Experimental_StdioMCPTransport), embeddings/reranking, provider setup, middleware, telemetry, and error handling. Use when building with generateText/streamText, generateObject/streamObject, tools (needsApproval, strict, inputExamples, activeTools, toolChoice, experimental_context), embeddings (embed/embedMany/rerank), or MCP tools/resources/prompts/elicitation.

Use built-in Codex coordination, retrieval, and utility tools effectively during live sessions. Trigger when work benefits from maintaining a step plan, asking the user 1-3 short multiple-choice questions before risky edits, inspecting a local image from disk, searching or browsing the web for current or source-backed information, discovering additional MCP tools or plugins, or running independent developer-tool calls in parallel.

Expo and React Native motion for iOS, Android, and web: Reanimated 4, worklets, gestures, transitions, Skia, accessibility, and manifest-driven validation.

Core Rust engineering. Use for implementation, debugging, review, compiler errors, ownership and lifetimes, traits and generics, async and concurrency, typed errors, Cargo and MSRV, crate choice, tests, performance, unsafe and security. Excludes focused product surfaces owned by narrower Rust skills.

Commit + push relevant files, split conventional commits, open PR to main with title/body. Triggers—commit, push, open PR, scaffold PR/release defaults.

Production-ready Supabase integration patterns for Next.js/React/TypeScript applications. Use when working with Supabase for (1) SSR authentication with @supabase/ssr, (2) Database operations and migrations, (3) Row Level Security (RLS) policies, (4) Storage buckets and file uploads, (5) Realtime channels and presence, (6) Edge Functions with Deno, (7) pgvector embeddings and semantic search, (8) Vercel deployment and connection pooling, (9) CLI operations (type generation, migrations). Triggers on Supabase client setup, auth patterns, RLS policies, storage uploads, realtime subscriptions, Edge Functions, vector search, or Vercel+Supabase deployment.

Independent gpt-5.6 diff review via the Codex CLI, normal or steerable adversarial with JSON findings. Use before shipping nontrivial changes.

Audit Convex—schema, security, runtime edges, migrations, function risk. Triggers—review, contract, remediate. Not greenfield spec (convex-feature-spec).

Use the Parallel CLI for web data work: URL extraction, deep research, data enrichment, entity discovery (findall), memory recall, and web monitoring. Web search goes to the parallel-search MCP server (free); escalate to `parallel-cli search` only when MCP is insufficient. Use for deep research, exhaustive investigation, extracting full content from URLs, enriching lists, finding entities, recalling prior runs, tracking web changes, and CLI setup/auth/balance.

Expert guidance for building AI agents with ToolLoopAgent (AI SDK v6+). Use when creating agents, configuring stopWhen/prepareStep, callOptionsSchema/prepareCall, dynamic tool selection, tool loops, or agent workflows (sequential, routing, evaluator-optimizer, orchestrator-worker). Triggers: ToolLoopAgent, agent loop, stopWhen, stepCountIs, prepareStep, callOptionsSchema, prepareCall, hasToolCall, InferAgentUIMessage, agent workflows.

OKLCH color space and color usage for web projects. Convert hex/rgb/hsl to oklch, generate palettes, check contrast, handle gamut boundaries, theme with Tailwind v4, and apply color with meaning. Triggers on oklch, color conversion, palette generation, contrast ratio, APCA, gamut, display p3, design tokens, semantic color tokens, hue drift, chroma, dark mode colors, accent color, color meaning, light and dark appearance, increased contrast.

Layout structure for web interfaces, from grouping and alignment to reading order, progressive disclosure, and adaptive breakpoints. Use when structuring a page or component, spacing or aligning controls, deciding what collapses at small sizes, handling RTL layout direction, or reviewing frontend code for layout. Triggers on layout, spacing, alignment, grouping, negative space, whitespace, visual hierarchy, reading order, progressive disclosure, breakpoints, responsive layout, container queries, safe area, full-bleed, edge-to-edge, layout margins, RTL layout, logical properties.

Adversarial design, frontend, and motion review for web or mobile branches. Use to inspect a dirty UI worktree, prove behavior in a real browser or device, measure motion, performance, and accessibility, and implement one canonical end state.

Explicit-only Kimi-powered frontend/UI orchestration skill and CLI for repo profiling, design-system context generation, Codex/Kimi Code/Claude Code adapter setup, plan-first git worktree runs, status/reply/finalize lifecycle control, and review artifacts. Use only when the user explicitly invokes `$kimi-ui-agent` or asks to use Kimi UI Agent for frontend/UI audits, redesigns, components, screenshots, implementation planning, or worktree orchestration.

Rust command-line apps with clap: derive and builder APIs, subcommands, flags, config and env precedence, stdout, stderr and JSON contracts, exit codes, completions, and binary distribution.

Explicit-only Rust architecture orchestrator. Use only when the user explicitly invokes rust-mega-eng for broad Rust ecosystem architecture, multi-crate workspace strategy, large refactors, release engineering, crate selection portfolios, or end-to-end Rust product planning across CLI, TUI, Tauri, services, libraries, CI, security, and distribution.

UX writing and interface copy, from button labels to error messages and empty states. Use when writing or reviewing any user-facing text: button and link labels, form errors, placeholders, settings labels, onboarding flows, notifications, or empty states. Docs and marketing prose belong to writing-guidelines. Triggers on UX writing, microcopy, interface copy, product copy, copywriting, button labels, link text, error messages, empty states, placeholder text, settings labels, capitalization, title case, sentence case.

Delegate implementation, investigation, or bulk work to gpt-5.6 codex via pinned codex exec. Use for clear-spec builds, migrations, debugging, or any task MODELS.md routes to codex.

TanStack Query v5: queryOptions/keys, caching, mutations, invalidation, optimistic updates, SSR hydration, and infinite queries.

TanStack Start: server functions/routes, middleware, SSR, auth, execution boundaries, and deployment for full-stack React apps.

Deep, cited research across official docs, Codex web tools, Context7 API, GitHub, package source, rendered web pages, and Firecrawl with evidence ledgers.

Pre-merge branch and PR closeout. Use to identify and implement follow-up work, hard cuts, dependency-native simplification, validation coverage, docs alignment, and merge-readiness fixes.

Sync repo docs (AGENTS, README, ADRs, specs, runbooks, doc comments) ↔ code/workflow. Triggers—big change, drift, docs-align/AGENTS prompts, plan, governance. Repo-native tools, any stack.

Repo/monorepo modernization: dependency upgrades, security fixes, deprecation cleanup, framework migrations, dependency-native refactors, and verified hard-cut simplification.

Implicit Rust Tauri v2 skill. Use for Tauri apps, Rust commands, invoke contracts, app state, plugins, capabilities and permissions, secure IPC, filesystem and shell access, sidecars, updater, bundling, desktop and mobile distribution, and Rust frontend bridge design.

TanStack Router: file routes, generated trees, typed root context, loaders/deps, validated search params, code splitting, and errors.

Architect AWS infrastructure and application integrations with production-grade service selection, boundaries, rollout thinking, and verification. Use when the user asks for AWS architecture, infra design, service selection, or AWS-backed implementation planning. Do not use for repo-specific Nova guidance unless the active repo adapter calls into it.

Cross-discipline interface review and build that coordinates better-accessibility, better-layout, better-writing, better-typography, better-colors and better-ui. Use for a holistic UI audit, a full interface review, or a cross-discipline design review of a screen, flow, feature, or product interface, or to build one with the same rules applied forward. Triggers on better-interface, full interface review, holistic UI audit, cross-discipline design review, review the whole interface. Modes: quick, core, full, build.

Compress prose-heavy docs into caveman format while preserving technical substance, code, paths, URLs, and structure.

Docker + Compose—arch, implement, harden, CI. Triggers—Dockerfile, compose, bake, dockerignore, audit, build/run debug, multi-svc dev/prod.

Fix unresolved GitHub PR review threads with codex-dev: fetch hosted state, verify, patch, validate, commit, push, and close out.

Art-directs an existing three.js or React Three Fiber scene to cinematic quality with postprocessing, HDRI and PBR lighting, contact shadows, transmission materials, and camera choreography. Use for look-development once the scene renders, not for scene setup or lifecycle.

Implicit Rust TUI and Ratatui skill. Use for terminal UIs, ratatui widgets and layouts, crossterm or termion event loops, terminal rendering architecture, async input, app state machines, keybindings, terminal UX, snapshot tests, and TUI performance.

Create, validate, install, and smoke-test Codex custom subagents from TOML templates. Use for ~/.codex/agents or project .codex/agents roles, subagent templates, model and sandbox defaults, or reusable agent packs.

Web typography from choosing fonts to spacing, wrapping and accessibility. Use when picking or pairing typefaces, configuring variable fonts or OpenType features, setting up a type scale, checking heading hierarchy, styling text in components, truncating text, styling underlines, selection, placeholders or carets, or reviewing frontend code for typography. Triggers on typography, fonts, font formats, woff2, variable fonts, font-weight, opentype, font-feature-settings, letter-spacing, line-height, type scale, heading hierarchy, heading levels, tabular numbers, text-wrap, truncation, line clamp, underlines, text-decoration, text selection, iOS input zoom, font smoothing, measure, line length, text-box, smart punctuation, drop cap.

Implicit Rust web service skill. Use for Axum, Tokio, Tower, Hyper, REST APIs, middleware/layers, request extractors, typed errors, tracing, config, graceful shutdown, SQLx, background jobs, service tests, and production Rust HTTP architecture.

Internal engineering docs: specs, ADRs, architecture, runbooks, migrations, rollout plans, and maintainer guides.

Dash Mantine Components (DMC) best practices and performance optimization guidelines. Use this skill when writing, reviewing, or refactoring Dash apps with DMC to ensure optimal patterns. Triggers on: DMC components, Mantine styling, Dash callbacks, dcc.Store, theming, or performance optimization in Plotly Dash applications.

Pre-PR multi-model review, parallel opus and codex exec adversarial lanes, then adversarial verification of merged findings. Read-only. Use before shipping nontrivial diffs.

Architect and implement modern full-stack web/native platform work across Next.js, Expo, Convex, and monorepo tooling. Use when the task needs stack detection, platform-level planning, cross-cutting architecture, or repo-native verification workflows. Route general web UI review to repo-local web-design-guidelines and React/Next performance review to repo-local vercel-react-best-practices when installed. Do not use for narrow dependency upgrades, PR review remediation, docs-only alignment, or Dash-only audits.