data-skills-plugin

data-skills-plugin

Claude Code plugin with two expert skills for Scala and Apache Spark development.

Skills

spark-scala

Expert guidance for Apache Spark development in Scala. Covers:

• SparkSession setup and configuration
• DataFrame vs Dataset API (with type-safe case classes)
• Transformations, actions, and lazy evaluation
• Schema definition and data sources (Parquet, Delta, JDBC, Kafka)
• Aggregations and window functions
• Join strategies (broadcast, sort-merge, skew handling)
• Partitioning, caching, and checkpointing
• RDDs and when to use them
• Structured Streaming with watermarks and stateful operations
• UDFs and when to avoid them
• Error handling: dead letter pattern, idempotent writes, corrupt data
• Clean transformation design and testing

Reference docs:

• references/optimization.md — AQE, Catalyst, shuffle tuning, anti-patterns
• references/streaming-patterns.md — watermarking, stateful ops, exactly-once
• references/data-sources.md — format selection, Parquet/Delta/JDBC/Kafka patterns
• references/error-handling.md — dead letter pattern, idempotent writes, checkpoint recovery
• references/rdd-api.md — when to use RDDs, PairRDD aggregation patterns, accumulators
• references/spark-ml.md — MLlib pipelines, transformers, estimators, evaluation

scala

Expert guidance for idiomatic, functional Scala. Covers:

• Pure functions, immutability, referential transparency
• Domain modeling: case classes, sealed trait ADTs, smart constructors
• Avoiding primitive obsession with opaque types
• Pattern matching (type, structure, guards, nested)
• Error handling: Option, Either, Try, ValidatedNel (no exceptions in pure code)
• Higher-order functions and immutable collections
• For comprehensions as monadic composition
• Tail recursion with @tailrec
• Type classes and given/using (Scala 3)
• Concurrency: Futures, ExecutionContext, Cats Effect IO
• Clean Code and Clean Architecture principles applied to Scala

Reference docs:

• references/fp-patterns.md — Functor, Monad, Monoid, IO monad, Validated, State
• references/collections.md — collection hierarchy, performance guide, idioms
• references/type-system.md — variance, bounds, higher-kinded types, opaque types
• references/domain-modeling.md — case classes, sealed traits/ADTs, smart constructors
• references/concurrency-futures.md — Futures, ExecutionContext, Cats Effect IO patterns
• references/testing.md — TDD, ScalaTest, property-based testing, test doubles
• references/clean-architecture.md — layers, dependency rule, use cases, ports & adapters
• references/code-smells.md — Scala-specific smells, refactoring recipes

Installation

claude plugin marketplace add drtey/scala-spark-plugin
claude plugin install data-skills-plugin@scala-spark-skills

Verify installation

Once installed, the skills are available as slash commands:

/spark-scala   → Apache Spark + Scala expert
/scala         → Idiomatic Scala expert

Structure

data-skills-plugin/
├── .claude-plugin/
│   ├── plugin.json
│   └── marketplace.json
└── skills/
    ├── spark-scala/
    │   ├── SKILL.md
    │   └── references/
    │       ├── optimization.md
    │       ├── streaming-patterns.md
    │       ├── data-sources.md
    │       ├── error-handling.md
    │       ├── rdd-api.md
    │       └── spark-ml.md
    └── scala/
        ├── SKILL.md
        └── references/
            ├── fp-patterns.md
            ├── collections.md
            ├── type-system.md
            ├── domain-modeling.md
            ├── concurrency-futures.md
            ├── testing.md
            ├── clean-architecture.md
            └── code-smells.md