Total 53,944 skills, Code Quality has 2431 skills
Showing 12 of 2431 skills
Verify code against paper. Use when user asks "does this match the paper", "check my implementation", or is implementing equations/algorithms from literature.
Systematically find blind spots in code, architecture, APIs, and deployment — structured critique that catches what familiarity hides
Standalone quality review for Elastic integrations. Classifies files by domain, loads domain-specific skills and review checklists, applies cross-domain consistency rules, CEL version verification, API conformance, and severity calibration. Input-agnostic: works on local packages, PR diffs, or branch comparisons. Use when reviewing integration quality independently of any build or fix workflow.
Create custom Code Analyzer rules for Regex (pattern matching), PMD (XPath/AST for Apex and metadata XML), and ESLint (LWC/JavaScript/TypeScript). Use when users want to enforce coding standards, ban patterns, detect hardcoded values, govern metadata, or add rules not in the built-in set. TRIGGER when: user says 'create a rule', 'ban System.debug', 'enforce naming convention', 'detect hardcoded IDs', 'custom rule', 'xpath rule', 'regex rule', 'add a PMD rule', 'enforce a policy', 'create a check for', 'flag this pattern', 'make a rule that catches', 'metadata rule', 'check permissions', 'enforce API version', 'eslint rule', 'lwc rule', 'override rule threshold', 'customize complexity', or describes a pattern to enforce. DO NOT TRIGGER when: user wants to run a scan (use dx-code-analyzer-run), configure engines (use dx-code-analyzer-configure), or explain existing rules (use dx-code-analyzer-run).
Hard-bug DIAGNOSIS + FIX: reproduce the failure, name the cause, write a regression test, apply the minimal fix. Use whenever the user reports a bug, failure, flaky test, perf regression, error, or visible misbehaviour whose cause is not yet known — even if they only paste a symptom, stack trace, or failing test output, or say "why is X broken", "it stopped working", "it got slower", "this looks wrong". Do NOT start debugging inline without this skill: if your next step would be forming hypotheses about an unexplained failure, invoke /pasteurize first. Do NOT use for review-only diffs (/age), feature design (/mold), fixes where the cause is already known (/cook), or when the user opted out of writes (/culture).
Apply the Tell Don't Ask (TDA) principle when reviewing, writing, or refactoring object-oriented code. Use this skill whenever the user asks about OOP design, mentions getters/setters, wants to review a class for encapsulation issues, asks how to move logic closer to data, or asks why code feels "procedural" despite using classes. Also trigger when the user asks to refactor code that queries an object's state before making decisions externally. This skill should kick in for any code review or design question involving data access patterns, encapsulation, or how objects should collaborate.
Review PyTorch pull requests for code quality, test coverage, security, and backward compatibility. Use when reviewing PRs, when asked to review code changes, or when the user mentions "review PR", "code review", or "check this PR".
This skill should be used when setting up code quality tooling with ESLint v9 flat config, Prettier formatting, Husky git hooks, lint-staged pre-commit checks, and GitHub Actions CI lint workflow. Apply when initializing linting, adding code formatting, configuring pre-commit hooks, setting up quality gates, or establishing lint CI checks for Next.js or React projects.
Apply the "How I Made Your Machine" coding style guide to implementation, refactoring, and code review tasks across TypeScript, Rust, and Python. Use when a request asks for this style guide, when improving maintainability and type safety, when modeling domain concepts with explicit variants/types, or when enforcing behavior-first testing.
Elevate a working PR: hindsight review, refactor, simplify, test audit, docs, quality gates. Composes: hindsight-reviewer agent, /refactor, /simplify, /update-docs, /check-quality, /distill. Use when: PR works but could be better. "How would we do this knowing what we know now?"
Diagnose and fix errors with Codex delegation. Traces error to root cause, researches approach, delegates fix, verifies. Use when: bug reports, error messages, stack traces, test failures.
Evaluate and improve code modularization using the Balanced Coupling Model. Analyzes coupling strength, connascence types, and distance to identify refactoring opportunities and architectural improvements. Use when reviewing code architecture, refactoring modules, or designing new systems.