Agent skills
Skills you can use with AI coding agents, indexed from public GitHub repositories.
-
config-management
Validate config at startup, secrets in memory only. Never read config during requests, never store secrets in env vars. Use node-env-resolver for multi-source config.
jagreehal/jagreehal-claude-skills 1
-
code-flow-analysis
Trace code execution path before implementing fixes. Forces understanding of fn(args, deps) flows, Result types, and workflow composition. Prevents guessing by requiring file:line references and execution diagrams.
jagreehal/jagreehal-claude-skills 1
-
investigation-modes
Three explicit investigation modes: LEARNING (build understanding), INVESTIGATION (diagnose problems), SOLVING (implement fixes). Prefix messages with mode, ask before transitioning.
jagreehal/jagreehal-claude-skills 1
-
fn-args-deps
Enforce the fn(args, deps) pattern: functions over classes with explicit dependency injection
jagreehal/jagreehal-claude-skills 1
-
debugging-methodology
Evidence-based debugging with Iron Law discipline. Instrument before guessing, trace before theorizing. Use when encountering any bug, test failure, or unexpected behavior - before proposing fixes.
jagreehal/jagreehal-claude-skills 1
-
git-worktrees
Use when starting feature work that needs isolation from current workspace. Creates isolated git worktrees with directory selection, safety verification, and baseline testing.
jagreehal/jagreehal-claude-skills 1
-
parallel-agent-dispatch
Use when facing 2+ independent tasks that can be worked on without shared state. Dispatch one agent per problem domain for concurrent investigation.
jagreehal/jagreehal-claude-skills 1
-
performance-testing
Load testing, chaos engineering, and performance validation. Prove your system works under pressure with k6, trace correlation, and progressive load profiles.
jagreehal/jagreehal-claude-skills 1
-
result-types
Never throw for expected failures. Use Result<T, E> types with explicit error handling and workflow composition.
jagreehal/jagreehal-claude-skills 1
-
strict-typescript
Enforce patterns with TypeScript beyond strict:true. Include noUncheckedIndexedAccess, erasableSyntaxOnly, ts-reset, and type-fest. Advanced type patterns and ESLint enforcement.
jagreehal/jagreehal-claude-skills 1
-
verification-before-completion
Use when about to claim work is complete, fixed, or passing. Requires running verification commands and confirming output before making any success claims. Evidence before assertions, always.
jagreehal/jagreehal-claude-skills 1
-
create-tasks
Creates well-formed tasks following a template that engineers can implement. Use when creating tasks, defining work items, creating tasks from PRD, breaking down features, or converting requirements into actionable tasks.
jagreehal/jagreehal-claude-skills 1
-
data-visualization
Visualization is communication. Chart selection, encoding hierarchy, accessibility, rendering performance. Use established algorithms - these problems are solved.
jagreehal/jagreehal-claude-skills 1
-
literal-answers
Treat questions as literal questions. Answer them honestly without interpreting as hidden instructions. STOP after answering - don't immediately act on assumed implications.
jagreehal/jagreehal-claude-skills 1
-
design-exploration
Use before any creative work - creating features, building components, adding functionality. Explores user intent, requirements, and design before implementation. One question at a time, multiple choice preferred.
jagreehal/jagreehal-claude-skills 1
-
concise-output
Enforce extreme brevity and high signal-to-noise ratio. Every word must justify its existence. Eliminates verbose explanations, filler phrases, and unnecessary elaboration.
jagreehal/jagreehal-claude-skills 1
-
pattern-enforcement
Enforce architectural patterns with ESLint rules. Block infra imports, enforce object params, prevent server/client leaks.
jagreehal/jagreehal-claude-skills 1
-
writing-tests
Principles for writing effective, maintainable tests. Covers naming conventions, assertion best practices, and comprehensive edge case checklists. Based on BugMagnet by Gojko Adzic.
jagreehal/jagreehal-claude-skills 1
-
structured-writing
Writing assistance with voice preservation. Structure content, ask clarifying questions, identify gaps. Small edits execute; large changes ask first.
jagreehal/jagreehal-claude-skills 1
-
research-first
Research-driven investigation. Validate solutions and explore documentation before presenting. Never ask questions you can answer yourself through research.
jagreehal/jagreehal-claude-skills 1
-
code-review-reception
Use when receiving code review feedback, before implementing suggestions. Requires technical verification, not performative agreement or blind implementation.
jagreehal/jagreehal-claude-skills 1
-
react-development
Modern React architecture patterns and web interface guidelines. Container/View split, framework adapters, React Query, dependency injection, Storybook-first development, accessibility, interactions, animations, performance. User experience over developer convenience.
jagreehal/jagreehal-claude-skills 1
-
spine-framework
Evaluate and improve technical articles using the SPINE framework. Use when writing blog posts, technical documentation, tutorials, or reviewing draft articles for quality.
jagreehal/jagreehal-claude-skills 1
-
tdd-workflow
Strict TDD state machine with Result types. 7-state workflow (PLANNING, RED, GREEN, REFACTOR, VERIFY, BLOCKED, VIOLATION). Every message MUST announce state. Integrates with fn(args, deps), vitest-mock-extended, and typed error handling.
jagreehal/jagreehal-claude-skills 1