Coding & Development

Ayuda a decidir qué plantilla(s) CLAUDE.md copiar a la carpeta de un proyecto concreto, sea de código (apps, scripts) o no (guión, producción audiovisual, escritura), usando como referencia la colección de archivos .md (plantillas, guía de migración, plantilla de historial, docs de plataforma) que el usuario tiene guardados en una carpeta local de su ordenador. Versión de esta skill para usar dentro de Claude Code, con acceso directo al sistema de archivos: lee las plantillas, lee el...

Ayuda a decidir qué plantilla(s) CLAUDE.md copiar a la carpeta de un proyecto concreto, sea de código (apps, scripts) o no (guión, producción audiovisual, escritura), usando como referencia la colección de archivos .md (plantillas, guía de migración, plantilla de historial, docs de plataforma) que el usuario tiene guardados en el Proyecto activo de Claude.ai. Actívala cuando el usuario pregunte por un proyecto concreto y quiera elegir, comparar o aplicar plantillas de CLAUDE.md para Claude...

Hand off an existing independent issue that surfaced mid-task to its own Claude Code session in a new cmux workspace, with auto-collected context; that session runs issue→worktree→PR alone. Not for splitting the current task.

> Make a slow test suite or CI test job faster and prove it — rank where the time goes on CI across many runs, prefer structural fixes (shared setup, parallel files, matrix legs) over per-file rewrites, probe parallel safety, and accept the change only on a measured CI before/after comparison.

> On-demand home for the pre-merge approval procedure — the three-surface pre-ask probe, grading every open finding by its blocking decoration, carrying the anchor's `Unverified` gaps into a follow-up, the six-part briefing, and the `AskUserQuestion` approve-ask that ends it.

Shared, versioned planning of a project in its repository - the backlog (docs/BACKLOG.md) where bugs, features, improvements and adjustments are captured for later and grouped under epics, the roadmap (docs/ROADMAP.md) of releases and the items each one delivers, and complete, uncompressed plans (docs/plans, feature or bug plans) with milestones, lineage, status history, promotion of backlog items, semantic search and drift detection through a local ChromaDB index. Use to capture or triage...

Deliver a PBI, Bug, Feature or Epic item by item with subagents (implement, review, PR, CI, merge). Use for "/deliver 12" or "deliver feature 7".

Orchestrator of the role-based development cycle - phases from product inception (Phase 0) and discovery to a verified release, on the project's branching model (integration, trunk or release-branches), the role skill called in each phase (product-inception, software-design, quality-assurance, devops-release, user-docs, architecture-docs, plan-memory, code review), a full path for features and a short path for small fixes, the project backlog with epics and bugs before new work, the roadmap...

Guides creation, audit, and improvement of portable Agent Skills using the shared SKILL.md format and Anthropic's established methodology, without confusing the portable specification with runtime-specific extensions. Use when the user wants to "create a skill", "create an agent skill", "build a skill", "make a new skill", "write a new skill", "improve an existing skill", "audit my skill", "audit a SKILL.md", "refine my skill", "test a skill", "test skill discovery", "install skills across...

Evaluate and use SSH Remote Compute before choosing where to run GPU, high-memory, parallel, batch, model-inference, bioinformatics, or other long-running scientific work; supports short remote commands and asynchronous jobs with automatic harvest and analysis.

>- Scans a codebase for structural quality issues (architecture incoherence, design-principle violations, dead code, excessive complexity, unclear naming, hidden state) and produces an evidence-backed, severity-ranked report with concrete, non-breaking refactor suggestions. Detects the stack and adds that stack's current, sourced rules as extra lenses where a stack pack exists (Swift/iOS, React). Each aspect gets its own reviewer on a capable model; a cheaper checker verifies and ranks every...

Run several independent Cotal features concurrently by creating one Git worktree and one spawn-capable mesh manager per feature; each manager staffs a review panel in a dedicated channel, adds one independent cold reviewer briefed under the cold-review skill, owns plan-to-commit delivery, and escalates unresolved product decisions by DM to the coordinator for relay to the user. Use when the user asks to split multiple features across managers/worktrees, run parallel feature teams on Cotal,...

Medir y mejorar la velocidad de una tienda PrestaShop (1.7, 8 y 9) y de su servidor sin romper nada. Copia y vuelta atrás antes de cada cambio, medición antes y después con navegador real (móvil en 4G), imágenes WebP (núcleo, plantilla, banners y CMS), recursos que bloquean el pintado (fuentes, iconos, jQuery, reCAPTCHA), caché del navegador, ajustes del servidor (OPcache, MariaDB, PHP-FPM, sesiones, DNS) y módulos d…

Türkçe metni insan eliyle yazılmış gibi doğal yazar veya onarır. Kullanıcı Türkçe bir metin için "insanileştir", "humanize et", "AI hissiyatını sil" veya "doğal Türkçeye çevir" dediğinde, metinde çeviri kokusu ya da TDK hatası denetimi istediğinde, Türkçe blog, makale, kurumsal yazı, sosyal medya içeriği veya teknik doküman yazarken kullan.

Design and implement embedded terminal rendering (cell grid, VT, fonts, sprites, paint cache). Use when building or improving a terminal pane, PTY viewer, agent TUI host, libghostty-vt / Ghostty-style paint, box drawing, Powerline glyphs, Nerd Font icons, or when terminal output looks wrong, soft, or misaligned compared to Ghostty / kero / herdr.

Generate and edit Obsidian Excalidraw drawings stored as Markdown (.excalidraw.md) — knowledge maps, decision trees, formula boards, teaching canvases. This skill should be used when a request involves creating or fixing an Excalidraw canvas inside an Obsidian vault, especially when it contains LaTeX formulas, needs theme/color alignment with existing notes, or when a previously generated canvas renders blank/washed-…

Make architecture, performance, and production decisions for serving generative AI models. Use when selecting models or hardware, diagnosing latency or throughput, optimizing LLM or multimodal inference, benchmarking, scaling GPU services, or estimating serving cost.

抖音电商（抖店 / 抖音小店 / 抖音商城）商品主图违规预审。对照抖音电商官方规则（商品主图发布规范、虚假宣传、价格违规、诱导第三方、类目行业规范），检查主图的画面、文字和规格，按 🔴明确违规 / 🟡经验风险 / ⚪可能误判 分级，指出问题位置和规则依据，并给出可以直接替换的合规文案。用户发来商品主图、详情图、SKU 图或主图文案，问能不能过审、会不会违规、为什么被驳回，或者请你检查、预审、优化时，都应使用本 skill，即使没有提到「违规」或「抖音」；也用于分析抖店驳回通知并把它记录成案例。

Compose, render and check a song in FL Studio from a text idea using the p5midi tool. Use when the user asks for a beat, song, loop, track or instrumental made in FL Studio, or wants an idea turned into an .flp / MIDI / MP3.

完整的 GEO（生成式引擎优化）服务流水线：给一个产品官网 URL 和介绍材料， 做站点诊断与 AI 答案采样、生成带验收标准的执行工单、产出可直接部署的资产 （llms.txt / JSON-LD / 定义块 / FAQ / 内容大纲与初稿）、自动验收工单是否闭环、 并打包成可直接发给客户的交付物。可按周期复跑，做长期 GEO 运营与月报。 国内和海外双市场并行：国内覆盖 智谱GLM/豆包/DeepSeek/Kimi/MiniMax/纳米AI/百度AI， 海外覆盖 Gemini/ChatGPT/Claude/Grok/Perplexity，问题库与指标按市场分开算。 当用户说「做 GEO」「生成式引擎优化」「让 AI 推荐我的产品」「AI 搜索里搜不到我们」 「GEO 方案/诊断/监测/月报/交付」「给客户做 GEO 服务」，或给一个产品网址要做 AI 可见性优化时使用。

Write, review or refactor Go code that is easy to read and still fast. Keeps functions small with low cyclomatic and cognitive complexity, puts performance into API design and data layout rather than clever code, and proves changes with benchmarks, allocation counts and escape analysis.

Reviews changes between HEAD and a fixed base (branch, commit, tag, or merge-base) on two separate axes: Standards (repository conventions, security, tests, and maintainability) and Spec (does the change implement what the issue, PRD, or spec requested?). Use when the user asks for a branch, PR or work-in-progress review or says "review since X".

使用基于 SSH 的 CLI 安全操作远端服务器并管理本地连接配置。适用于列出连接、远程执行命令、上传下载文件、启动或停止可视化配置编辑器，以及确认参数、返回值、配置路径和环境校验步骤的场景。

Reduce unnecessary test runs and agent context usage while preserving required verification. Use when planning, running, or improving verification for code changes, or when optimizing test reporting and logs.