Claude-Code-Unified-Skill-Plugin-Library

Benchmark compensation against market data. Trigger with "what should we pay", "comp benchmark", "market rate for", "salary range for", "is this offer competitive", or when the user needs help evaluating or setting compensation levels.

Answer questions about company policies, benefits, and procedures. Trigger with "what's our policy on", "how does PTO work", "benefits question", "expense policy", "remote work policy", or any question about company rules, perks, or procedures.

Analyze workforce data — attrition, engagement, diversity, and productivity. Trigger with "attrition rate", "turnover analysis", "diversity metrics", "engagement data", "retention risk", or when the user wants to understand workforce trends from HR data.

Plan and execute organizational or technical changes. Trigger with "we're changing", "rolling out", "migration plan", "how do we communicate this change", "change management plan", or when the user is planning a change that affects people, processes, or systems.

Plan and optimize resource allocation. Trigger with "resource planning", "capacity", "utilization", "staffing plan", "who should work on what", "we're stretched thin", or when the user needs help allocating people, budget, or time across projects and teams.

Evaluate, compare, and manage vendor relationships. Trigger with "evaluate this vendor", "compare vendors", "vendor review", "should we renew", "RFP", or when the user is making procurement or vendor decisions.

Write effective microcopy for user interfaces. Trigger with "write copy for", "help with UX copy", "what should this button say", "error message for", "empty state copy", or when the user needs help with any interface text.

Create comprehensive developer handoff documentation from designs. Trigger with "handoff to engineering", "developer specs", "implementation notes", "design specs for developers", or when a design needs to be translated into detailed implementation guidance.

Manage design tokens, component libraries, and pattern documentation. Trigger with "design system", "component library", "design tokens", "style guide", or when the user asks about maintaining consistency across designs.

This skill should be used when the user's animations feel slow, choppy, or frame-drop. Trigger when the user mentions "animation is janky", "animation is choppy", "dropped frames", "stuttering animation", "animation lags", "smooth on desktop but janky on mobile", "animation causes layout shifts", "scroll is not smooth", "60fps fail", "animation slows down page", "will-change", "GPU acceleration", "how to profile animation", "DevTools performance panel", "Chrome performance tab", or asks why their animation feels slow despite looking correct in code. Also trigger when the user asks about optimizing or auditing existing animations for performance.

This skill should be used when the user needs to tune spring physics for a specific feel, or wants to understand and configure Framer Motion spring parameters. Trigger when the user mentions "spring animation", "stiffness and damping", "Framer Motion spring", "iOS spring feel", "feels too bouncy", "feels too stiff", "feels too slow", "no overshoot", "spring preset", "mass and damping", "inertia animation", "drag spring", "motion values", "useSpring hook", or asks how to replicate the feel of native iOS or Android springs. Also trigger when the user says an animation "doesn't settle right", "oscillates too much", or wants a "snappier" or "smoother" spring.

This skill should be used when the user wants to implement gesture-driven interactions or motion that responds to touch or pointer input. Trigger when the user mentions "swipe to dismiss", "drag to reorder", "pinch to zoom", "long press", "drag gesture", "touch event", "pan gesture", "flick or fling motion", "momentum-based animation", "spring on release", "pull to refresh", "swipeable card", "draggable element", or wants an interaction to "feel like iOS", "feel native", or "track my finger". Also trigger when the user asks how to handle gesture velocity, inertia, or dismiss thresholds.

This skill should be used when the user needs production-ready animation patterns for specific UI states or feedback moments. Trigger when the user mentions "loading state", "skeleton screen", "loading spinner", "error animation", "error shake", "success animation", "success checkmark", "form feedback", "button loading state", "checkbox animation", "toggle animation", "notification badge", "like animation", "copy to clipboard feedback", "progress indicator", "toast notification animation", or wants to add "polish" or "juice" to individual UI components. Also trigger when the user asks how to animate specific component states (empty state, error state, success state, pending state).

This skill should be used when the user needs to make animations safe and compliant for users with motion sensitivities, or wants to audit existing animations for accessibility. Trigger when the user mentions "prefers-reduced-motion", "vestibular disorder", "motion sensitivity", "WCAG animation", "WCAG 2.3.1", "seizure threshold", "flashing content", "parallax accessibility", "dizzy from animation", "animation causes nausea", "motion a11y", "reduce motion", "animation accessibility audit", or asks how to make their animations WCAG compliant. Also trigger when the user is working on a medical, government, or high-compliance product and mentions any animation, or when accessibility is mentioned alongside motion design.

This skill should be used when the user wants to add animations, transitions, motion, or movement to any UI. Trigger when the user mentions "add animation", "page transitions", "hover effects", "loading states", "scroll effects", "parallax", "spring animation", "shared element transitions", "gesture-driven motion", "SVG animation", "kinetic typography", or "3D depth effects". Also trigger when the user says an interface "feels static", "feels lifeless", "feels janky", "feels too abrupt", "feels slow", or wants it to "feel like Apple", "feel like iOS", "feel like Vercel", "feel like Linear", "feel smooth", or "feel polished". Use this skill whenever motion is anywhere in the conversation — motion is a design system, not a feature.

This skill should be used when the user wants animations that are triggered by or linked to scroll position. Trigger when the user mentions "scroll animation", "scroll-triggered", "on scroll", "scroll reveal", "fade in on scroll", "parallax", "horizontal scroll", "sticky section", "scroll-linked animation", "pin section on scroll", "scroll progress", "Intersection Observer", "GSAP ScrollTrigger", "useScroll Framer Motion", or wants elements to animate as the user scrolls. Also trigger when the user asks about "hero animation", "scroll storytelling", or mentions a landing page needs to feel "alive while scrolling".

This skill should be used when the user needs to orchestrate multiple elements animating in sequence or with complex timing relationships. Trigger when the user mentions "stagger animations", "animation sequence", "choreograph animations", "elements animate one after another", "staggered reveal", "delay between animations", "GSAP timeline", "Framer Motion variants", "animation timing", "orchestrate", "50ms stagger", or has a component where multiple elements need to animate in a specific order. Also trigger when the user says an animation "looks like everything explodes at once" or wants a more "coordinated" or "sequential" entrance.

Audit designs and code for WCAG 2.1 AA compliance. Trigger with "is this accessible", "accessibility check", "WCAG audit", "can screen readers use this", "color contrast", or when the user asks about making designs or code accessible to all users.

Evaluate designs for usability, visual hierarchy, consistency, and adherence to design principles. Trigger with "what do you think of this design", "give me feedback on", "critique this", "review this mockup", or when the user shares a design and asks for opinions.

This skill should be used when the user asks about "neural architecture search", "NAS", "architecture design", "model selection", "Pareto analysis", "proxy metrics", "zero-cost NAS", "search space design", "DARTS", "ENAS", "EfficientNet", "architecture efficiency", "FLOP count", "parameter count", "model capacity", or when deciding which model family to use for a given task and constraint set.

This skill should be used when the user asks about "distributed training", "multi-GPU training", "data parallelism", "model parallelism", "pipeline parallelism", "tensor parallelism", "DDP", "FSDP", "ZeRO", "DeepSpeed", "Megatron-LM", "GPU utilization", "NCCL", "torchrun", "gradient communication", "checkpoint recovery", "spot instance preemption", "NVLink", "InfiniBand", "training throughput", or when a model is too large for a single GPU or training speed needs to be scaled.

This skill should be used when the user asks about "AI governance", "model fairness", "bias detection", "algorithmic bias", "disparate impact", "SHAP", "LIME", "model explainability", "interpretability", "XAI", "differential privacy", "DP-SGD", "federated learning", "adversarial robustness", "FGSM", "model compliance", "EU AI Act", "GDPR Article 22", "responsible AI", "model audit", "regulatory compliance", or when deploying AI in regulated industries (finance, healthcare, hiring, criminal justice).

This skill should be used when the user asks about "generative models", "diffusion models", "DDPM", "DDIM", "stable diffusion", "GANs", "GAN training", "generator", "discriminator", "mode collapse", "FID score", "language models", "LLM", "GPT", "transformer", "text generation", "image generation", "ELBO", "score matching", "latent diffusion", "variational autoencoder", "VAE", "CLIP", "multimodal generation", or when building any system that generates novel data samples.

This skill should be used when the user asks about "hyperparameter tuning", "HPO", "Bayesian optimization", "Gaussian process", "acquisition function", "Expected Improvement", "Upper Confidence Bound", "Hyperband", "ASHA", "Optuna", "Ray Tune", "learning rate search", "grid search", "random search", "hyperparameter importance", "fANOVA", or when a model's validation loss is suboptimal and systematic tuning is needed.

This skill should be used when the user asks about "loss functions", "gradient descent", "backpropagation", "vanishing gradients", "regularization", "bias-variance tradeoff", "statistical learning theory", "PAC learning", "VC dimension", "overfitting", "underfitting", "cross-entropy loss", "KL divergence", "ELBO", "maximum likelihood estimation", "Bayesian inference", or any foundational mathematical concepts underlying machine learning models and optimization.

This skill should be used when the user asks about "MLOps", "ML pipeline", "data pipeline", "feature engineering", "feature store", "data preprocessing", "model deployment", "model serving", "model registry", "experiment tracking", "MLflow", "Weights and Biases", "model versioning", "CI/CD for ML", "model monitoring", "data quality", "schema validation", "reproducibility", "technical debt in ML", or when operationalizing a machine learning model for production.

This skill should be used when the user asks about "reinforcement learning", "RL", "reward function", "policy gradient", "PPO", "SAC", "DQN", "Q-learning", "actor-critic", "MDP", "Markov decision process", "environment design", "reward shaping", "exploration strategy", "experience replay", "multi-agent RL", "RLHF", "reward hacking", or when training an agent to interact with an environment to maximize cumulative reward.

This skill should be used when the user asks about "concept drift", "data drift", "model degradation", "model staleness", "production model monitoring", "self-healing", "automatic retraining", "shadow model", "champion-challenger", "blue-green deployment", "canary deployment", "Kolmogorov-Smirnov test", "PSI", "Population Stability Index", "model refresh", "continuous learning", "online learning", or when a production model's performance has degraded over time due to distribution shift.

This skill should be used when the user asks about "data pipeline", "ETL", "ELT", "data engineering", "Apache Airflow", "Prefect", "Dagster", "dbt", "Spark", "Kafka", "streaming pipeline", "batch pipeline", "data ingestion", "data transformation", "data quality", "data validation", "Great Expectations", "feature store", "feature engineering", "feature pipeline", "data lake", "data warehouse", "medallion architecture", "bronze silver gold", "Parquet", "Iceberg", "Delta Lake", "dbt model", "pipeline orchestration", "data lineage", "backfill", "incremental processing", "change data capture", "CDC", or "real-time data". Also trigger for "my pipeline is slow", "data is duplicated", "pipeline is failing", or "how do I build a data pipeline for ML".

This skill should be used when the user asks about "machine learning", "deep learning", "neural network", "train a model", "PyTorch", "TensorFlow", "JAX", "scikit-learn", "XGBoost", "LightGBM", "fine-tune", "transfer learning", "model architecture", "loss function", "optimizer", "learning rate", "batch size", "epoch", "overfitting", "underfitting", "regularization", "dropout", "batch normalization", "gradient descent", "backpropagation", "training loop", "validation", "hyperparameter tuning", "Optuna", "Ray Tune", "Weights & Biases", "MLflow", "model checkpoint", "early stopping", "mixed precision", "distributed training", "GPU training", "CUDA", "model serving", "TorchServe", "ONNX", or "model deployment". Also trigger for "my model isn't converging", "loss is NaN", "training is slow", "model is overfitting", or "how do I improve my model accuracy".

This skill should be used when the user asks about "evaluate a model", "model evaluation", "evaluation metrics", "accuracy", "precision", "recall", "F1 score", "AUROC", "AUC-ROC", "confusion matrix", "BLEU", "ROUGE", "BERTScore", "model bias", "fairness metrics", "slice analysis", "error analysis", "calibration", "reliability diagram", "expected calibration error", "A/B test model", "shadow deployment", "champion-challenger", "offline evaluation", "online evaluation", "benchmark", "regression metrics", "MAE", "RMSE", "R squared", "model performance", "why is my model bad", or "production model monitoring". Also trigger for "how do I know if my model is good", "my model is worse in production than offline", or "how to measure fairness".

This skill should be used when the user asks about "prompt engineering", "prompt design", "system prompt", "few-shot examples", "chain of thought", "CoT", "zero-shot", "one-shot", "few-shot", "prompt optimization", "prompt iteration", "improve accuracy of LLM", "make the model follow instructions", "structured output", "JSON mode", "function calling", "tool calling", "prompt template", "LangChain", "LlamaIndex", "prompt injection", "jailbreak defense", "RAG prompt", "retrieval augmented generation prompt", "evaluation of prompts", or "why is the LLM not doing what I want". Also trigger for "my prompt gives inconsistent results", "the model ignores my instructions", or "how do I write a better system prompt".

This skill should be used when the user asks about "API authentication", "API authorization", "JWT", "JSON Web Token", "access token", "refresh token", "token rotation", "OAuth 2.0", "OAuth flow", "authorization code flow", "client credentials", "PKCE", "API key", "Bearer token", "session authentication", "cookie authentication", "httpOnly cookie", "CSRF", "CORS with credentials", "token expiry", "token refresh", "invalidating tokens", "logout", "revocation", "stateless auth", "stateful auth", "auth middleware", "validate JWT", "sign JWT", "decode JWT". Also trigger for "how do I protect my API routes", "how do I implement login", "should I use JWT or sessions", "how to handle token expiry", or "users being logged out too often".

This skill should be used when the user asks about "GraphQL", "GraphQL schema", "GraphQL query", "GraphQL mutation", "GraphQL subscription", "resolvers", "type definitions", "SDL", "schema definition language", "DataLoader", "N+1 in GraphQL", "GraphQL pagination", "GraphQL fragments", "directives", "federation", "Apollo", "Relay", "schema stitching", "schema federation", "persisted queries", or "GraphQL vs REST". Also trigger for "how do I design this GraphQL type", "how to batch GraphQL requests", "why are my GraphQL queries slow", or "how to paginate in GraphQL".

This skill should be used when the user asks about "REST API", "API design", "endpoint design", "HTTP methods", "status codes", "API versioning", "CRUD endpoints", "RESTful", "API conventions", "pagination", "cursor pagination", "offset pagination", "resource naming", "URL design", "API specification", "OpenAPI", "Swagger", "API contract", "request body", "response schema". Also trigger for "how should I design this API", "what status code should I use", "how to version an API", "is this good API design", "review my endpoint", or "how to structure my routes".

Review code for bugs, security vulnerabilities, performance issues, and maintainability. Trigger with "review this code", "check this PR", "look at this diff", "is this code safe?", or when the user shares code and asks for feedback.

This skill should be used when the user asks about "database migration", "schema migration", "ALTER TABLE", "adding a column", "dropping a column", "renaming a column", "changing column type", "adding an index", "dropping an index", "zero-downtime migration", "blue-green migration", "expand-contract", "backfill", "lock contention", "table lock", "migration strategy", "forward migration", "rollback migration", "Flyway", "Liquibase", "Alembic", "Prisma migrate", "Rails migration", "Django migration". Also trigger for "how do I safely rename a column", "how to add NOT NULL constraint without downtime", "my migration is locking the table", or "apply migration without downtime".

This skill should be used when the user asks about "slow query", "query optimization", "EXPLAIN", "EXPLAIN ANALYZE", "query plan", "sequential scan", "index scan", "index-only scan", "bitmap scan", "N+1 query", "JOIN performance", "subquery vs JOIN", "CTE performance", "window functions", "GROUP BY performance", "ORDER BY performance", "LIMIT OFFSET", "pagination query", "full-text search", "query rewrite", "missing index", "unused index", "covering index", "partial index", "composite index", "statistics", "autovacuum", "table bloat", "VACUUM", "ANALYZE", "pg_stat_statements", "slow log". Also trigger for "this query is slow", "how can I speed up this query", "my database is slow", "query takes too long", or "how to avoid N+1".

This skill should be used when the user asks about "database schema", "data modeling", "table design", "column types", "primary key", "foreign key", "indexes", "normalization", "denormalization", "database relationships", "one-to-many", "many-to-many", "entity relationship", "ER diagram", "database constraints", "NOT NULL", "unique constraint", "CHECK constraint", "soft delete", "deleted_at", "audit log", "audit table", "JSONB", "JSON column", "enum type", "PostgreSQL schema", "MySQL schema", "Prisma schema", "SQLAlchemy models", "ActiveRecord migration". Also trigger for "how should I store this data", "how do I model this relationship", "is this database design good", or "how to structure the database for".

This skill should be used when the user asks about reading, searching, or analyzing logs. Trigger on "logs", "log message", "log file", "grep logs", "log parsing", "log aggregation", "structured logs", "JSON logs", "log level", "debug log", "error in logs", "warning in logs", "LogQL", "Loki", "Elasticsearch query", "Kibana", "Splunk", "Datadog logs", "Cloudwatch Logs", "log correlation", "trace ID", "request ID", "log shipping", "log rotation", "log sampling", "high-cardinality logging". Also trigger for "I see this error in logs", "what does this log message mean", "how do I find related log entries", or "why is there a spike in errors".