r-error-handling
jjjermiahBase R error handling with tryCatch, withCallingHandlers, and custom condition classes. Use when implementing error recovery, debugging conditions, or working with stop/warning/message—e.g., "tryCatch in R", "custom condition class", "handle warnings and errors", "error recovery patterns".
python-polars
jjjermiahEnforces Polars over Pandas for functional pipe-style data manipulation (like dplyr in R). Use when writing Python data processing code, data transformation pipelines, ETL workflows, or analytical queries—e.g., "process this CSV", "aggregate sales data", "filter and transform DataFrame", "group by and calculate metrics".
python-code-quality
jjjermiahPython code quality with Ruff linting/formatting and ty type checking. Use when configuring Ruff rules, setting up ty type checking, writing pyproject.toml quality config, creating pixi quality tasks, enforcing type annotations, or fixing lint errors—e.g., "set up ruff and ty", "configure Python linting", "add type checking to project", "fix ruff violations".
r-rlang-programming
jjjermiahrlang metaprogramming for tidy evaluation and non-standard evaluation (NSE) in R. Use when building data-masking APIs, wrapping dplyr/ggplot2/tidyr functions with {{, !!, !!! operators, implementing quosures and dynamic dots, or designing tidyverse-style DSLs — e.g., "tidy eval wrapper function", "embrace operator usage", "NSE programming patterns", "custom select helpers".
pixi-tasks
jjjermiahComplex pixi task workflows and orchestration. Use when building task dependency chains, configuring caching with inputs/outputs, creating parameterized tasks, or setting up CI pipelines—e.g., "pixi task depends-on", "task caching for build automation", "multi-environment test matrices".
pixi-expert
jjjermiahComprehensive pixi package manager skill for all pixi operations from beginner to advanced. Use for initializing projects, managing dependencies, configuring environments, multi-environment setups, workspace composition, system requirements (CUDA/glibc), task workflows, CI/CD integration, or any pixi.toml/pyproject.toml configuration — e.g., pixi init, pixi add numpy, setup multi-environment project, configure CUDA, monorepo workspace, pixi.lock issues, Docker with pixi, GitHub Actions pixi.
code-reviewer
jjjermiahProvides structured code review against plans and standards. Use when a feature is complete and needs validation, when reviewing code before merge, or when assessing quality and test coverage—e.g., "finished step X", "ready for review", "validate architecture", "check quality and tests".
python-production-libs
jjjermiahProduction-grade Python library selection and recommendations. Use when choosing libraries for HTTP clients, CLI frameworks, data validation, structured logging, JSON serialization, terminal output, or async patterns—e.g., "which library for HTTP", "modern alternatives to requests", "pydantic vs dataclasses", "structured logging setup".
r-benchmarking
jjjermiahR benchmarking, profiling, and performance analysis with reproducibility and measurement rigor. Use when timing R code execution, profiling with Rprof or profvis, measuring memory allocations, comparing function performance, or optimizing bottlenecks — e.g., "benchmark R function", "profvis profiling", "microbenchmark comparison", "performance analysis", "memory profiling".
azuredevops-pipelines-logging
jjjermiahAzure DevOps Pipelines logging-command guidance for reliable script-to-agent signaling, variable passing, and log UX. Use when writing or debugging `##vso[...]` and `##[...]` commands in YAML/Bash/PowerShell pipelines, troubleshooting output variable scope, handling secrets and masking behavior, or publishing summaries/artifacts from scripts. Pair with `azuredevops-pipelines-template` when template architecture and logging semantics are both in scope.
r-documentation
jjjermiahR package documentation with roxygen2 and Rd files, including mathematical notation, selective Rd parsing, and structured sections. Use when writing/updating/refactoring roxygen2 documentation, adding math formulas to R help pages, programmatically reading Rd files, or troubleshooting Rd rendering
script-writer
jjjermiahWrite production-ready one-off scripts and automation utilities with proper error handling and safety patterns. Use when developing bash automation, Python CLI tools, shell scripts, system administration scripts, or command-line batch processing—e.g., "write a script to process files", "python one-liner for data conversion", "bash automation for backups", "shell script with error handling".
azuredevops-pipelines-template
jjjermiahAzure DevOps Repos-first template architecture for reusable CI pipelines. Use when designing or debugging Azure Pipelines YAML templates, splitting PR validation from post-merge main workflows while reusing one core CI template, enforcing strict compile-time/runtime variable rules, building typed template APIs (stepList, jobList, stageList, templateContext), or creating self-contained dependency-management templates for TypeScript/Python/R using bun, npm, and pixi—e.g., "one CI template for PR and main", "fix expression timing bug", "design jobList template contract".
r-testing
jjjermiahR package testing with testthat (3rd edition). Use when writing R tests, fixing failing tests, debugging errors, or reviewing coverage—for example: "write testthat tests", "fix failing R tests", "snapshot testing", "test coverage".
python-pybytesize
jjjermiahByte size parsing and formatting with pybytesize. Use when converting bytes to human-readable sizes, parsing size strings, or doing block-aligned calculations—e.g., "format 1GB as MiB", "parse '500MB'", "human readable file size".
r-expert
jjjermiahCore R programming skill for all R code, package development, and data science workflows. Use when writing R functions, building packages, using tidyverse (dplyr, ggplot2, purrr), creating Shiny apps, working with R Markdown/Quarto, or doing data analysis—e.g., "write an R function", "refactor this R code", "create a Shiny dashboard", "set up package tests", "debug R errors".
skill-creator
jjjermiahGuides creation, validation, and packaging of AI agent skills with token-efficient design, progressive disclosure patterns, and YAML frontmatter best practices. Use when building new skills, updating existing skills, validating skill structure against standards, or packaging for distribution—e.g., "create skill", "validate SKILL.md", "package skill for sharing", "check description format".
Python testing
jjjermiahPytest-first Python testing with emphasis on fakes over mocks. Covers unit, integration, and async tests; fixture design; coverage setup; and debugging test failures. Use when writing tests, reviewing test quality, designing fixtures, setting up pytest, or debugging failures—e.g., "write unit tests for new feature", "fixture design patterns", "fakes vs mocks comparison", "fix failing tests".