All authors
Claude Skills by orrrrli
github.com/orrrrli19 skills0 installs4 views
- Android Clean ArchitectureClean Architecture patterns for Android and Kotlin Multiplatform projects — module structure, dependency rules, UseCases, Repositories, and data layer patterns. Use when structuring modules, layers, or data flow in an Android or KMP project.Votes: 0GitHub stars: 2
- Audit Layer BoundariesScan the codebase for the three architectural violations that break the layer contract in this project. Use when adding new features, after a large refactor, or whenever you suspect imports have drifted.Votes: 0GitHub stars: 2
- DiagnoseDisciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.Votes: 0GitHub stars: 2
- Dotnet Clean ArchitectureScan a .NET Clean Architecture solution for the five violations that break the dependency rule — Domain referencing EF Core, handlers touching DbContext directly, Application referencing Infrastructure, controllers bypassing MediatR, and exceptions used for expected business errors. Use when adding a feature, after a large refactor, or whenever project references have drifted.Votes: 0GitHub stars: 2
- Fill ContextBuild or regenerate the full .claude/ context tree by scanning the codebase and grilling the developer on what can't be inferred from code alone. Produces a multi-file context structure (business, architecture, domains, engineering) with the root CLAUDE.md as the index. Use after project init or whenever context is missing, stale, or incomplete.Votes: 0GitHub stars: 2
- Improve Codebase ArchitectureScan a codebase for deepening opportunities, informed by the project's .claude/ context tree and past decisions in engram, present them as a claude.ai Artifact report, then grill through whichever one the user picks.Votes: 0GitHub stars: 2
- Prd To Github BacklogTurn a PRD/spec into a GitHub Project with a full engineering backlog: epics → story sub-issues → task sub-issues, organized by phase and sprint, with sizes/estimates, area labels (backend/frontend/design/QA/security/devops), and project fields (Status/Sprint/Size/Estimate). Use when the user wants to create a GitHub project + issues + sub-tasks from a PRD, build/seed a backlog, decompose a spec into epics/stories/tasks, or set up a project board from requirements.Votes: 0GitHub stars: 2
- Project BootstrapScaffold a brand-new full-stack project in one pass — backend .NET Clean Architecture + CQRS (MediatR) and mobile React Native/Expo — going from an already-filled root CLAUDE.md context straight to a working skeleton plus exactly one vertical slice example feature implemented end-to-end. Fills the gap between fill-context (context only, no code) and spec-apply (per-issue, too granular for day-zero). Use when the user wants to "levantar el proyecto", bootstrap a new app from scratch, scaffold ...Votes: 0GitHub stars: 2
- Prompt RewriteAudit or rewrite LLM instructions — a skill (SKILL.md), CLAUDE.md, agent definition, system prompt, tool description or a single prompt — against Anthropic's prompting guidance for current Claude models. Diagnoses first, then proposes the rewrite with the reason for each change; it does not overwrite the file until the user approves. Use when the user asks to improve, audit, tighten or fix a prompt or skill, or asks why an agent keeps doing the wrong thing (skips a stop, over-builds, ignores ...Votes: 0GitHub stars: 2
- Spec ApplyImplement a GitHub Story and all its Task sub-issues using a spec-driven loop: the AC become failing tests written by test-writer, then ponytail code makes them pass: resolve → (per task: explore → propose → failing tests [test-writer] → apply [with ponytail] → verify [+ mutation check] → commit → archive) → PR. Use when the user provides a Story issue number, a Story ID like 'BE-AUTH-1' from a backlog generated by prd-to-github-backlog, a single Task ID like 'BE-AUTH-1-T1', or says 'implemen...Votes: 0GitHub stars: 2
- Spec VerifyVerify an implementation against its Story's acceptance criteria and the code-quality spec, in a fresh code-reviewer subagent. Delivers PASS/FAIL per criterion with file:line evidence, after running the project's own tests. Report only: never fixes, never closes issues. Use after implementing a task, when the user says "verify this", "does this match the spec", "spec-verify", and from spec-apply Phase 4.Votes: 0GitHub stars: 2
- TddTest-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.Votes: 0GitHub stars: 2
- Write SwiftHow to write modern Swift well — modeling with value types, Swift 6 data-race safety and approachable concurrency (@concurrent, main-actor-by-default, actors, task groups), protocols and generics (some vs any), API design, performance and ARC, Swift Testing, macros, and the modern language features agents don't know about yet. Use when writing, reviewing, or migrating Swift, or when a concurrency error, a hang, a data race, a retain cycle, or a performance problem needs fixing.Votes: 0GitHub stars: 2
- Blind PickMake the options for a hard-to-reverse decision fight before picking one. One advocate sub-agent per option argues for it and attacks the others, then three judges who do not know which option was recommended score them on a rubric that weighs fit with the whole feature and long-term cost over the smallest diff. Use when the user says "blind-pick", "make the options fight", or doubts a recommended option; or right after options were presented for a schema, API contract or architecture decisio...Votes: 0GitHub stars: 2
- CavemanUltra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.Votes: 0GitHub stars: 2
- Grill MeInterview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".Votes: 0GitHub stars: 2
- HandoffGenerate a structured handoff document for the current session — what was done, current state, pending work, blockers, and next steps — and copy it to the OS clipboard to paste into the next chat. Writes no file (unlike claude-mem:handoff, which writes HANDOFF.md). Use when the user says /handoff or wants to hand the session to another chat or person via clipboard.Votes: 0GitHub stars: 2
- Apple DesignApple's design rules for iOS screens you can check in a static design file — translucent materials and depth, typography (text styles, weight, leading, Dynamic Type), accessibility variants (Reduce Transparency, Increase Contrast), and Apple's eight design principles with their tactical rules (feedback, wayfinding, grouping, labels). Use when designing or reviewing an iOS screen in a design tool such as pen.dev or Figma, or when asked whether a screen "feels like iOS". For motion and gestures...Votes: 0GitHub stars: 2
- Swiftui MotionBuild motion in a SwiftUI app (iOS 17+), making the decisions in the order that decides whether it feels right — should it animate at all, what purpose, which API, which spring, how it interrupts, how a drag hands off to the animation, which haptic, and how it degrades under Reduce Motion. Writes the implementation with stock SwiftUI only. Use when animating a view, adding a transition, press feedback, a custom drag or swipe, a haptic, or when motion feels sluggish, bouncy or disconnected. Fo...Votes: 0GitHub stars: 2