Deep Learning

Package management for modular code organization. This skill should be used when creating, installing, or modifying metool packages, working with package structure conventions, or adding Claude Code skills to packages.

Testing patterns with Bun test runner, coverage thresholds, mocking, and CI/CD integration. Use when writing tests, organizing test files, or setting up quality gates.

Design, build, and maintain comprehensive knowledge bases. Bridges document-based (RAG) and entity-based (graph) knowledge systems. Use when building knowledge-intensive applications, managing organizational knowledge, or creating intelligent information systems.

3層開示モデル（メタデータ→本文→リソース）による段階的な情報提供で、トークン効率と知識スケーラビリティを両立。スキル発動信頼性を最大化し、必要な時に必要な知識だけをロードします。Anchors:• The Pragmatic Programmer (Andrew Hunt, David Thomas) / 適用: 段階的な情報開示と実践的改善 / 目的: トークン効率を維持しながら深い知識を提供• Progressive Disclosure (Jakob Nielsen) / 適用: 認知負荷の最小化 / 目的: UX設計原則のスキルメタデータへの応用• Information Architecture (Louis Rosenfeld) / 適用: 階層的知識組織化 / 目的: 遅延読み込みとインデックス駆動設計Trigger:Use when designing skill metadata, optimizing token usage, implementing progressive disclosure patterns, improving skill activation reliability, organizing knowledge hierarchically, reducing context window consumption, or creating scalable documentation structures.

優先順位付けフレームワークの専門スキル。MoSCoW法、RICE Scoring、Kano Modelを用いて、限られたリソースで最大の価値を提供するための意思決定を支援します。Anchors:• 『Inspired』（Marty Cagan） / 適用: プロダクト優先順位 / 目的: 価値最大化• Intercom RICE Scoringガイド / 適用: 定量的スコアリング / 目的: データドリブンな意思決定• Kano Model理論（Noriaki Kano） / 適用: 顧客満足度分析 / 目的: 戦略的投資判断Trigger:Use when prioritizing features, requirements, backlog items, or strategic initiatives. Apply to sprint planning, release planning, roadmap development, feature evaluation, or resource allocation decisions.

This skill should be used when the user asks to "create CLAUDE.md", "initialize CLAUDE.md", "sync CLAUDE.md with code", "update documentation from codebase", "create .claude/rules/", "manage modular rules", "split large CLAUDE.md", or mentions project documentation setup. Manages complete lifecycle of project instructions including creation, synchronization with code patterns, and modular organization.

Frontend development guidelines for React/TypeScript applications. Modern patterns including Suspense, lazy loading, useSuspenseQuery, file organization with features directory, styling best practices, routing, performance optimization, and TypeScript best practices. Use when creating components, pages, features, fetching data, styling, routing, or working with frontend code.

Maintain Claude Code memory hygiene by auditing, organizing, updating, and optimizing memory files in `.claude/memory/`. Use when users request memory cleanup, organization, updates, or want to reduce context pollution. Handles stale content, redundancy, conflicts, and file organization issues.

Generate a coverage report for all localization files using the project's check-i18n tool. Use when auditing translation completeness or identifying languages needing attention.

Generate comprehensive sensor health and status reports across all LimaCharlie organizations. Use when users ask about sensor connectivity, data availability, offline sensors, sensors not reporting events, or fleet-wide health queries (e.g., "show me sensors online but not sending data", "list sensors offline for 7 days across all orgs").

Organize design assets, optimize images and fonts, maintain brand asset libraries, implement version control for assets, and enforce naming conventions. Keep design assets organized and production-ready.

This skill optimizes deep learning models using various techniques. It is triggered when the user requests improvements to model performance, such as increasing accuracy, reducing training time, or minimizing resource consumption. The skill leverages advanced optimization algorithms like Adam, SGD, and learning rate scheduling. It analyzes the existing model architecture, training data, and performance metrics to identify areas for enhancement. The skill then automatically applies appropriate optimization strategies and generates optimized code. Use this skill when the user mentions "optimize deep learning model", "improve model accuracy", "reduce training time", or "optimize learning rate".

Procedural noise functions in GLSL—Perlin, simplex, Worley/cellular, value noise, FBM (Fractal Brownian Motion), turbulence, and domain warping. Use when creating organic textures, terrain, clouds, water, fire, or any natural-looking procedural patterns.

This skill allows Claude to construct and configure neural network architectures using the neural-network-builder plugin. It should be used when the user requests the creation of a new neural network, modification of an existing one, or assistance with defining the layers, parameters, and training process. The skill is triggered by requests involving terms like "build a neural network," "define network architecture," "configure layers," or specific mentions of neural network types (e.g., "CNN," "RNN," "transformer").

SECIモデル（野中郁次郎）に基づく組織知識の形式知化と共有を専門とするスキル。暗黙知（経験、勘、ノウハウ）を形式知（ドキュメント、パターン）に変換し、体系化することで組織全体で再利用可能な知識として活用する。Anchors:• The Knowledge-Creating Company (Nonaka/Takeuchi) / 適用: SECIサイクル4フェーズ（共同化・表出化・連結化・内面化） / 目的: 暗黙知の特定・言語化・統合の理論的基盤• The Pragmatic Programmer (Hunt/Thomas) / 適用: 実践的改善とDRY原則 / 目的: 品質維持と重複知識の統合判断• Design Patterns (Gang of Four) / 適用: パターン記述形式 / 目的: 再利用可能な知識の抽象化と構造化Trigger:Use when formalizing tacit knowledge, documenting best practices, converting code review insights to reusable knowledge, managing organizational knowledge base quality, or applying SECI model workflows.Keywords: knowledge management, tacit knowledge, explicit knowledge, SECI model, documentation, best practices, pattern extraction

This skill detects hierarchical (nested) TAD structures from Hi-C contact maps (in .cool or mcool format) using OnTAD, starting from multi-resolution .mcool files. It extracts a user-specified chromosome and resolution, converts the data to a dense matrix, runs OnTAD, and organizes TAD calls and logs for downstream 3D genome analysis.

Author and improve Agent Skills following the agentskills.io specification. Use when creating new SKILL.md files, modifying existing skills, reviewing skill quality, or organizing skill directories with proper naming, descriptions, and progressive disclosure.

Expert guidance for building Entity Component System (ECS) applications with Sparze, a Zig ECS library. Use when working with Sparze ECS code for (1) Writing system functions with query filters, (2) Organizing systems with single responsibility and proper execution order, (3) Designing component architectures and groups, (4) Using query modifiers (Optional, Exclude, Free), (5) Managing resources and events, (6) Understanding performance trade-offs between Query/Group/SingleQuery, (7) Implementing event-driven system chains, (8) Implementing deferred commands pattern, or (9) Any other Sparze ECS development tasks.

Create custom slash commands for Claude Code. Use when the user wants to create, generate, or build a slash command (.md file) for Claude Code, or when they ask about slash command syntax, frontmatter options, argument handling, or want help organizing their commands. Triggers on phrases like "make a slash command", "create a command for Claude Code", "write a /command", or questions about $ARGUMENTS, allowed-tools, or command frontmatter.

This skill allows Claude to construct and configure neural network architectures using the neural-network-builder plugin. It should be used when the user requests the creation of a new neural network, modification of an existing one, or assistance with defining the layers, parameters, and training process. The skill is triggered by requests involving terms like "build a neural network," "define network architecture," "configure layers," or specific mentions of neural network types (e.g., "CNN," "RNN," "transformer").