giip-fde-agent

Audit and upgrade an EXISTING website or app to premium quality without a rewrite. Use when the user asks to redesign, revamp, refresh, polish, improve, modernize, or de-slop an existing frontend or codebase — it identifies generic AI design patterns and applies high-end standards while preserving functionality and the current stack (any CSS framework or vanilla CSS). Triggers: redesign, revamp, upgrade UI, improve design, polish existing site, make it look better, modernize, 리디자인, 리뉴얼, 개선, 다듬기, 손보기, リデザイン, 改善, 刷新. Do NOT use for building a NEW interface from scratch — use the frontend-design skill for that.

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.

Carve guardrail-adjacent items out of scope with safe alternatives before risk-adjacent work starts, then run the safe remainder at full strength in a fresh subagent that only ever sees the carved prompt, never the risky input. Use when a task includes stealth, scraping, privacy, IP, policy, licensing, security, or other safety-adjacent material that could be silently dropped, over-elaborated, or needlessly diluted. Fires on the impulse, not only the topic: the moment you notice yourself about to hedge, soften, silently skip, or brace for a refusal, carve before you execute.

Run repeated build -> QA -> retro -> re0-work cycles while preserving learning and letting code die. Use for long agentic projects where progress must be measured by quality-cleared templates, reusable modules, and eliminated anti-patterns rather than hours spent or features accumulated.

Audit any eval, metric, experiment, or benchmark for leakage — does external ground-truth enter independently, or are the model, scorer, and designer just confirming a result no outside truth ever produced? Use before trusting any 'how we'll know it worked' — an A/B, a holdout, a score, a validation — and whenever a result feels too clean or self-confirming. Walks an 8-pattern leakage taxonomy and returns only the patterns that fire, each with an independence fix. Read-only.

Read the live cycle state and return the single highest-leverage next best action, not a menu. Use when a project is between phases, the author asks what to do next, too many valid threads are open, or the work needs re-entry into frame, build, drive, retro, hate, re0-work, or ship.

Rewrite a finished commit's message into a clean, handoff-ready form in your own log style, so `git log` alone tells the story. User-invoked: run it after a commit.

Restart a project or artifact from v0 while preserving only proven lessons, contracts, gates, vocabulary, real-surface tests, and negative corpus. Use when the foundation is wrong, accumulated code is misleading progress, or a new pass should learn from the old one without inheriting its accidental architecture.

Cold-read the artifact you're focused on from a fresh, zero-context sub-session to confirm it stands on its own — a shower-thought reset for accumulated session bias. Use when a long session has worn away your fresh eyes and you can no longer tell whether the artifact in focus is clear to someone with no prior context; before a handoff, publish, or merge; or when you want a clean-room comprehension smoke test. Spawns a separate context-free reviewer; it diagnoses, it does not fix.

After you create or change an artifact or skill, taste-test it with our own skills instead of trusting your in-session judgment — recursive self-improvement, made automatic. Use right after writing or editing anything, before calling it done, committing, or handing it off.

Audit where one fact lives in many places and report the canonical source of truth plus a consolidation plan — read-only. Use when the same fact, spec, decision, config, status, or definition appears in multiple artifacts or platforms and may have drifted; when refactoring a legacy codebase, building or grading the severity of a knowledge base, or scaffolding a new project; before trusting a value stated in more than one place; or when asked to find duplication, check consistency, or locate the source of truth. Never modifies anything.

Consolidate a fact that's scattered across places into one canonical source and replace the rest with references — mutating. Use when the user asks to deduplicate, consolidate, unify, or establish a single source of truth across artifacts or platforms; or when the same fact lives in many places and copies are drifting. Folds unique details into the canonical home and reconciles contradictions before removing any copy.

Bounded text-space gradient optimization for SKILL.md prompts using execution traces. Loads low-reward traces (reward < 0.8) from .agent/traces/, groups them by skill, and applies bounded edits (max 15% word change) with a Validation Gate to prevent destructive rewrites. Use when you want to run the SkillOpt algorithm to automatically improve skill prompts, or when a skill repeatedly underperforms and you want a systematic text-space optimization pass. Triggers: skillopt, skill optimization, bounded edits, prompt gradient, validation gate, improve skill

Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies

로컬 파일 검색 시 jikji find를 사용하여 LLM 토큰 소비와 반복 호출을 줄이는 파일 탐색 스킬. 파일, 폴더, 메타데이터, 파싱된 문서 텍스트를 미리 구성된 에이전트 맵/검색 인덱스를 통해 효율적으로 찾습니다. Use proactively when local file/folder/document discovery is needed under an explicit root. Do NOT start with grep, ls, find, rg, tree, or cat for file location tasks. Triggers: find file, search file, locate file, file discovery, 파일 찾기, 파일 검색, 파일 탐색, ファイル検索, ファイル発見, 文件搜索, 文件查找, buscar archivo, localizar archivo, trouver fichier, recherche fichier, Datei suchen, Datei finden Do NOT use for: code quality review (use code-review), security audit (use gstack-security).

Use Open Design (nexu-io/open-design) — the open-source agent-native design workspace — to generate prototypes, dashboards, decks, images, and motion graphics. Apply brand-grade DESIGN.md systems and connect via MCP to the agent already in use. Use this skill when the user asks to: - Create a UI prototype, landing page, or dashboard - Generate a slide deck or presentation - Apply a specific brand or design system (Stripe, Vercel, Apple, etc.) - Set up Open Design's MCP integration with their agent - Export designs to HTML, PDF, PPTX, or MP4 - Produce motion graphics or video artifacts via HyperFrames Triggers: open design, od mcp, design workspace, design prototype, DESIGN.md system, 오픈 디자인, 디자인 워크스페이스, 프로토타입 생성, 디자인 시스템 적용, オープンデザイン, デザインワークスペース, プロトタイプ生成, デザインシステム適用

Codex의 로컬 상태를 점검하고 정리하여 성능을 유지하는 스킬입니다. 오래된 채팅·워크트리·로그·데드 프로젝트 참조가 쌓여 Codex가 느려질 때 사용합니다. 다음 상황에서 사용합니다: - Codex가 오래 사용한 후 무거워졌을 때 - 오래된 채팅·터미널·로그·워크트리가 누적되었을 때 - 정기 유지보수(주간/격주)가 필요할 때 - 아카이브 전 핸드오프 문서가 필요할 때 트리거 키워드: keep-codex-fast, codex 느림, codex 정리, codex 유지보수, codex maintenance, codex slow, codex cleanup, codex handoff, 코덱스 속도, 코덱스 정리 출처: https://github.com/vibeforge1111/keep-codex-fast 갱신 주기: 월 1회 이상 원본 레포 변경 사항 확인 권장 (auto-check 워크플로우: codex-maintenance.md)

Post-task retrospective and learning extraction skill. Inspired by gstack's /retro and integrated with K-Layer. Use this after completing a significant task or resolving a complex issue to document lessons learned and improve future performance. Triggers: retro, retrospective, post-mortem, lessons learned, 회고, 사후 분석, 振り返り

Guides the usage of the Google Gemini API with the Gen AI SDK. Covers common tasks like client initialization, multimodal usage (images, video, audio, files), chat, function calling, structured output, and safety settings. Supports Python, JS/TS, Go, Java, and C#.

Master the Gemini Interaction artifacts. Use this when a user wants to know about the Interaction artifacts or if you want to know about the Interaction artifacts yourself.

Guides the usage of the Gemini Live (Multimodal Live) API using the Gen AI SDK. Use when the user asks about real-time, bidirectional streaming of voice and video with Gemini. Covers configuration, session management, sending/receiving audio/video media, and tool integration.

Guides the usage of Gemini API on Google Cloud Vertex AI with the Gen AI SDK. Use when the user asks about using Gemini in an enterprise environment or explicitly mentions Vertex AI. Covers SDK usage (Python, JS/TS, Go, Java, C#), capabilities like Live API, tools, multimedia generation, caching, and batch prediction.

투자/트레이딩 관련 오픈소스·에이전트 스택을 객관적으로 평가하고, GIIP Agent System의 role/rule/skill/workflow/k-layer에 안전하게 이식할 때 사용합니다. 다음 상황에서 사용합니다: - 투자/트레이딩용 에이전트 구조를 설계하거나 개선할 때 - 특정 레포를 도입할지 비교·검토할 때 - 백테스트/리스크/거버넌스 체크리스트가 필요할 때 트리거 키워드: 투자, 트레이딩, invest, trading, quant, backtest, risk, investment analysis

Discovery and integration of professional design systems from multiple platforms (designmd.ai, designmd.app, getdesign.md, designmd.me). Use this when starting a project, updating UI, or requesting specific brand "looks" (Stripe, Vercel, etc.).

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

Code review skill for analyzing code quality, detecting bugs, and ensuring best practices. Provides comprehensive code review with actionable feedback. Use proactively when user requests code review, quality check, or bug detection. Triggers: code review, review code, check code, analyze code, bug detection, 코드 리뷰, 코드 검토, 버그 검사, コードレビュー, バグ検出, 代码审查, 代码检查, revisión de código, revisar código, detección de errores, revue de code, réviser le code, détection de bugs, Code-Review, Code überprüfen, Fehlererkennung, revisione del codice, rivedere codice, rilevamento bug Do NOT use for: design document creation, deployment tasks, or gap analysis (use phase-8-review).

Detects the current operating system, shell, and available tools to select the best execution strategy.

에이전트 작업 결과로부터 자동으로 source-linked claim을 생성하고, 지식 위키(K-Layer)에 축적합니다. 카파시 다이어그램 패턴: Raw Sources → LLM Processing → Curated Wiki → Outputs 다음 상황에서 사용합니다: - 반복 작업에서 패턴을 발견했을 때 (예: 특정 API 오류가 반복됨) - 실험/시도의 결과를 지식으로 전환할 때 - 이전 작업의 교훈을 다음 작업에 반영할 때 - "/k-layer" 명령어가 명시적으로 언급될 때 트리거 키워드: k-layer, knowledge, 지식, 학습, 패턴 발견, 교훈, claim, wiki

Performs a rapid security sweep for secrets, API keys, and common vulnerabilities using cross-platform platform tools.

Self-audits the agent's skills folder to ensure metadata completeness, valid frontmatter, and current documentation.

Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.

Core rules for bkit plugin. PDCA methodology, level detection, agent auto-triggering, and code quality standards. These rules are automatically applied to ensure consistent AI-native development. Use proactively when user requests feature development, code changes, or implementation tasks. Triggers: bkit, PDCA, develop, implement, feature, bug, code, design, document, 개발, 기능, 버그, 코드, 설계, 문서, 開発, 機能, バグ, 开发, 功能, 代码, desarrollar, función, error, código, diseño, documento, développer, fonctionnalité, bogue, code, conception, document, entwickeln, Funktion, Fehler, Code, Design, Dokument, sviluppare, funzionalità, bug, codice, design, documento Do NOT use for: documentation-only tasks, research, or exploration without code changes.

PDCA document templates for consistent documentation. Plan, Design, Analysis, and Report templates with proper structure. Use proactively when generating PDCA documents to ensure consistent format. Triggers: template, plan document, design document, analysis document, report, 템플릿, 계획서, 설계서, 분석서, 보고서, テンプレート, 計画書, 設計書, 模板, 计划书, 设计书, plantilla, documento de plan, documento de diseño, documento de análisis, informe, modèle, document de plan, document de conception, document d'analyse, rapport, Vorlage, Plandokument, Designdokument, Analysedokument, Bericht, modello, documento di piano, documento di progettazione, documento di analisi, rapporto Do NOT use for: code implementation, deployment, or non-documentation tasks.

Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.

Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.

Claude Code learning and education skill. Teaches users how to configure and optimize Claude Code settings. Works across any project and any language. Start learning/setup with "learn" or "setup". Use proactively when user is new to Claude Code, asks about configuration, or wants to improve their Claude Code setup. Triggers: learn claude code, claude code setup, CLAUDE.md, hooks, commands, skills, how to configure, 클로드 코드 배우기, 설정 방법, Claude Code 학습, クロードコード学習, 设置方法, how do I use claude code, aprender claude code, configuración, cómo configurar, apprendre claude code, configuration, comment configurer, Claude Code lernen, Konfiguration, wie konfigurieren, imparare claude code, configurazione, come configurare Do NOT use for: actual coding tasks, debugging, or feature implementation.

Desktop app development guide for cross-platform desktop applications. Covers Electron and Tauri frameworks. Use proactively when user wants to build desktop applications with web technologies. Triggers: desktop app, Electron, Tauri, mac app, windows app, 데스크톱 앱, デスクトップアプリ, 桌面应用, aplicación de escritorio, app de escritorio, application de bureau, application desktop, Desktop-Anwendung, Desktop-App, applicazione desktop, app desktop Do NOT use for: web-only projects, mobile apps, or server-side development.

9-phase Development Pipeline complete knowledge. Use when user doesn't know development order or starts a new project from scratch. Use proactively when user asks about development order, phases, what to do first, or starts a new project without clear direction. Triggers: development pipeline, phase, development order, where to start, what to do first, how to begin, new project, 개발 파이프라인, 뭐부터, 어디서부터, 순서, 시작, 開発パイプライン, 何から, どこから, 开发流程, 从哪里开始, pipeline de desarrollo, fase, orden de desarrollo, por dónde empezar, qué hacer primero, pipeline de développement, phase, ordre de développement, par où commencer, que faire d'abord, Entwicklungspipeline, Phase, Entwicklungsreihenfolge, wo anfangen, was zuerst tun, pipeline di sviluppo, fase, ordine di sviluppo, da dove iniziare, cosa fare prima Do NOT use for: ongoing implementation, existing feature work, or bug fixes.

Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.

Fullstack development skill using bkend.ai BaaS platform. Covers authentication, data storage, API integration for dynamic web apps. Project initialization with "init dynamic" or "dynamic init". Use proactively when user needs login, database, or backend features without managing servers. Triggers: fullstack, BaaS, bkend, authentication, login feature, signup, database, web app, SaaS, MVP, init dynamic, dynamic init, 풀스택, 인증, 로그인 기능, 회원가입, 데이터베이스, 웹앱, フルスタック, 認証, ログイン機能, データベース, 全栈, 身份验证, 登录功能, autenticación, inicio de sesión, registro, base de datos, fullstack, aplicación web, authentification, connexion, inscription, base de données, fullstack, application web, Authentifizierung, Anmeldung, Registrierung, Datenbank, Fullstack, Web-App, autenticazione, accesso, registrazione, database, fullstack, applicazione web Do NOT use for: static websites, Enterprise-grade systems requiring custom infrastructure.