cuda-guide

>

Migrate an apollo-react canvas component from Emotion styled-components to apollo-wind Tailwind classes following the BaseNode reference migration patterns

Scheduler configuration and the overnight session queue — task gates (`git_commit`, `git_diff`, `command`), `schedule` field reference (duration vs calendar, `at`/`every`/`after`/`delay`/`cooldown`/`not_before`/`not_after`/`except`), priority/pipeline ordering, one-time/max_runs tasks, workflow-backed tasks (`workflow:` + `inputs:`), overnight prepare-and-dispatch workflow. Use when adding or debugging scheduled tasks in `~/.agentwire/scheduler.yaml`, wiring up overnight sessions, or explaining how gating/dispatch works.

Complete guide for developing Klytos CMS plugins including structure, entry points, MCP tools, admin pages, hooks, routes, and best practices. Use when creating, modifying, extending Klytos functionality, adding MCP tools, admin pages, hooks, filters, or debugging plugins.

Streaming, pagination, count strategies, bulk operations

Acting as template repo maintainer: review proposals in `localdocs/proposals/` and optionally apply accepted ones to the upstream template.

Create or update Claude skills optimized for Skillmark benchmarks. Use for new skills, skill scripts, references, benchmark optimization, extending Claude's capabilities.

Guide for defining and using Claude subagents effectively. Use when (1) creating new subagent types, (2) learning how to delegate work to specialized subagents, (3) improving subagent delegation prompts, (4) understanding subagent orchestration patterns, or (5) debugging ineffective subagent usage.

This skill helps you add new rules to the attgo-linter project.

Guides how to integrate the NNews NuGet package for consuming the NNews CMS API in a .NET 8 project. Use when the user wants to consume articles, categories, tags, images, or AI-powered content generation from the NNews API.

Transform Claude into an expert code reviewer following industry best practices.

Draft or update design documents based on approved objectives. Use when translating requirements into technical specifications. Covers architecture, data model, API design, and design completeness assessment.

Execution-oriented workflow for HyperParallel distributed operator development. Analyzes the operator, implements or updates code and tests.

TypeScript-first schema validation library with static type inference and runtime validation

Generate a concise Product Requirements Document (PRD) from a feature description

Orchestrate Agent Teams for parallel multi-session collaboration. Use for research, implementation, review, and debug workflows requiring independent teammates.

Consult and apply the repository architecture reference when discussing system architecture, module boundaries, directory layout, service interfaces, Docker Compose, Kubernetes deployment, or operational entry points. Use when a user asks about architecture, refactors, project structure, deployment models, or infra layout.

Plans the next sprint with PM analysis, multi-agent technical table, and roadmap generation. Use to plan sprints, prioritize features, convene technical tables.

Phase 2: Generate a task list from requirements (PRD, CRD, or DRD). Creates actionable steps for implementation.

Expert knowledge for defining REST and WebSocket APIs with `schematic-define`. Use when creating or editing `RestApi`/`Endpoint` models, auth and env mappings, request or response shapes, and OpenAPI extension behavior that drives generated clients.

Self-improvement loop triggered after any user correction. Captures mistakes as prevention rules in tasks/lessons.md so the same error is never repeated.

Implementa feature nel sistema di fatturazione italiana validando contro normativa fiscale. Usa per aggiungere calcoli IVA, ritenuta d'acconto, split payment, imposta di bollo, gestione fatture PA, regime forfettario, numerazione progressiva, note di credito, o qualsiasi logica che deve rispettare DPR 633/72, DPR 600/73, DPR 642/72. NON usare per bug fix tecnici, refactoring, o modifiche UI senza impatto fiscale.

This is an AI-first project: any contributor — human or model — should be able to land a change that respects the architecture without needing to read the whole codebase first. The rules below are che

Draft a detailed Non-Disclosure Agreement between two parties covering information types, jurisdiction, and clauses needing legal review. Use when creating confidentiality agreements or preparing an NDA for a partnership.

배포 스크립트의 보안·안전성 리뷰를 수행한다. 롤백 절차, 장애 대응, 권한 설정을 점검한다.

Confidence-based model tier escalation for agent tasks. Defines when to escalate from Haiku to Sonnet to Opus, escalation signals, and the never-skip-tiers rule. Use when an agent fails or underperforms.

Post-draft diagnostic for sermon and book writing. Scans for machine tells, assesses authenticity risk, checks voice continuity against the author's sermon corpus. Fires after writing is complete or when reviewing existing drafts. For during-writing standards, see like-a-human.

Scan the codebase and auto-generate architecture map and conventions

Improve the eval system quality -- add better cases, fix broken cases, improve the LLM judge, add new metrics, or refine pass/fail criteria based on eval run data. Use when the user wants to make the eval system itself better.

>

Meta-level analysis of plan vs execution to identify process improvements. Analyzes divergence patterns and suggests updates to CLAUDE.md, agents, and skills.

nextjs-performance

>-

Load this skill when writing tests, reviewing test coverage, setting up testing strategies, or ensuring code quality. Covers the testing pyramid, unit/integration/e2e test standards, naming conventions, and mandatory 90% coverage requirements.

>-

startproject

Run a weekly retrospective across Threads and note outputs, then propose next-step improvements. Use for recurring optimization.

SoNash project context injected into agent definitions via skills: field

Product management: PRDs, RICE prioritization, metrics

This skill identifies and prevents common testing anti-patterns that make test suites unreliable, slow, and difficult to maintain. It provides detection strategies and refactoring solutions for health

>

Run unit and integration tests

通过 Context7 MCP 获取库和框架的实时最新文档，而非依赖训练数据，防止 API 幻觉。适用于查询任何库或框架的用法、配置、示例代码。触发词：怎么用、怎么配置、API参考、文档、示例代码、用法、接口、库文档、框架文档、documentation、docs、how to use、API reference、setup、configure、React怎么用、Next.js配置、Prisma查询、Vue用法、Express路由、Tailwind类名、Supabase认证、TypeScript类型、Zod验证、shadcn组件、Drizzle ORM、tRPC、Fastify、NestJS、Astro、SvelteKit、Nuxt、Vite、Vitest、Playwright。

Automatically scan code directories and generate or update structured Markdown documentation (index + detail) for Dart, Rust, TypeScript, JavaScript, React/Next.js and more. Use when you need to create or incrementally update API/type/component docs directly from source code.

zero-skills

>

>

>

>