Parcourir le registre

Toutes les entrées, par ordre alphabétique — 57 265 au total. Page 144 sur 230.

• large-file-kpi-analysis根据数据量自动选择读取策略（大文件转Parquet），提取关键指标进行单位一致性验证与排序分析，并输出可下载的结果表格。
• large-file-parquet-analysis-and-highlight当Excel文件总行数超过1万行时，通过转换为Parquet格式提升读取性能，提取目标指标并计算最大值，最后将结果输出为Excel并对特定行进行高亮标注。
• large-file-refactorPRB's collection of agent skills
• large-payload-completenessVerify large payload fetches are COMPLETE — anti silent-truncation. Cross-check size vs expected, follow all p
• large-repo-refactor在大型存量代码库做重构时使用。控制影响面,小步推进,不破坏现有行为。
• large-requirements-processor遵循ruoyi-vue-plus框架规范下自主开发的claude工具集
• large-workspace-handlingTo partition large workspaces (100+ files) into scoped subagent tasks when context is insufficient.
• largest-first-context-sheddingWhen a single API-level user message trips the per-message tool-result budget, drop the largest fresh results
• lark-approval飞书审批：查询和处理审批待办/已办/实例，搜索可发起审批定义、查看定义详情并发起原生审批实例。当用户要处理审批任务、查看审批实例、搜索或发起审批时使用。审批待办不是飞书任务；非审批类待办走 lark-task。不负责创建
• lark-apps妙搭（Spark/Miaoda）应用开发与托管：应用创建、本地全栈开发、云端生成迭代、创意设计（UI mockup / 可交互原型 / 线框图 / 落地页 / 仪表盘 / 幻灯片 deck / 视觉探索）、AI相关能力和
• lark-attendance飞书考勤打卡：查询自己的考勤打卡记录
• lark-base飞书多维表格（Base）操作：建表、字段、记录、视图、统计、公式/lookup、表单、仪表盘、workflow、角色权限；遇到 Base/多维表格/bitable 或 /base/ 链接时使用。文件导入/导出转 lark
• lark-calendar飞书日历：管理日历日程和会议室。查看/搜索日程、创建/更新日程、管理参会人、查询忙闲和推荐时段、预定会议室。当用户需要查看日程安排、创建/修改会议、查询/预定会议室时使用。不负责：查询过去的视频会议记录（走 lark-v
• lark-contact飞书 / Lark 通讯录:按姓名 / 邮箱解析成 open_id,或按 open_id 反查姓名 / 部门 / 邮箱 / 联系方式 / 个人状态 / 签名,以及按关键词搜索当前用户可见的机器人 / 智能体(agent)
• lark-doc飞书云文档（Docx / Wiki 文档）：读取和编辑飞书文档内容。当用户给出文档 URL 或 token，或需要查看、创建、编辑文档、插入或下载文档图片附件时使用。文档中嵌入的电子表格、多维表格、画板，先用本 skil
• lark-drive飞书云空间（云盘/云存储）：管理 Drive 文件和文件夹，包含上传/下载、创建文件夹、复制/移动/删除、查看元数据、查询权限设置、评论/权限/订阅、标题、版本、飞书文档密级标签（secure labels）和本地文件导
• lark-eventLark/Feishu real-time event listening / subscribing / consuming: stream events as NDJSON via `lark-cli event c
• lark-im飞书即时通讯：收发消息和管理群聊。发送和回复消息、搜索聊天记录、管理群聊成员、上传下载图片和文件（支持大文件分片下载）、管理表情回复、发送应用内/短信/电话加急、发送和处理交互卡片（Interactive Card）、监
• lark-mail飞书邮箱：Use when user mentions 起草邮件、写邮件、草稿、发送/回复/转发邮件、查阅邮件、看邮件、搜索邮件、邮件文件夹、邮件标签、邮件联系人、监听新邮件、邮件收信规则等；use for mail/e
• lark-markdown飞书 Markdown：查看、创建、上传、编辑和比较 Markdown 文件。当用户需要创建或编辑 Markdown 文件、读取、修改、局部 patch 或比较差异时使用。不负责将 Markdown 导入为飞书在线文档，
• lark-minutes飞书妙记：搜索妙记、查看妙记基础信息、下载/上传音视频、读取或编辑妙记的产物内容、改标题、替换说话人/关键词、申请妙记查看/编辑权限。当给出minute_token、本地音视频文件，要查/改/转妙记产物，或用户明确要主动
• lark-note飞书会议纪要（Note）直查：已知 note_id 时查询纪要详情、展示类型、关联文档 token，并读取 unified 原始逐字记录。当用户已持有 note_id，或从文档显式 vc-node-id 获得 note_
• lark-okr飞书 OKR：管理目标与关键结果。查看和编辑 OKR 周期、目标、关键结果、对齐关系、量化指标和进展记录。当用户需要查看或创建 OKR、管理目标和关键结果、查看对齐关系时使用。不负责：待办任务管理（lark-task）、
• lark-openapi-explorer飞书/Lark 原生 OpenAPI 探索：从官方文档库中挖掘未经 CLI 封装的原生 OpenAPI 接口。当用户的需求无法被现有 lark-* skill 或 lark-cli 已注册命令满足，需要查找并调用原生飞书
• lark-sharedUse for lark-cli setup/auth tasks: auth login/status/logout, user vs bot identity, business-domain permissions
• lark-sheets飞书电子表格：创建和操作电子表格。支持创建表格、管理工作表与行列结构（增删/合并/调整尺寸/隐藏/冻结）、读写单元格（值/公式/样式/批注/单元格图片）、查找替换、多操作原子批量更新，以及图表、透视表、条件格式、筛选器、
• lark-skill-maker创建 lark-cli 的自定义 Skill。当用户需要把飞书 API 操作封装成可复用的 Skill（包装原子 API 或编排多步流程）时使用。
• lark-slides飞书幻灯片：创建和编辑幻灯片。创建演示文稿、读取幻灯片内容、管理幻灯片页面（创建、删除、读取、局部替换）。当用户需要创建或编辑幻灯片、读取或修改单个页面时使用。当用户给出 doubao.com 的 /slides/ UR
• lark-task飞书任务：管理任务、清单和任务智能体。创建待办任务、查看和更新任务状态、拆分子任务、组织任务清单、分配协作成员、上传任务附件、注册或注销任务智能体、更新任务智能体的主页数据、写入智能体任务记录。当用户需要创建待办事项、查
• lark-vc飞书视频会议：搜索历史会议记录、查询会议纪要（总结/待办/章节/逐字稿）、查询参会人快照。当用户查询已结束的会议、获取会议产物（纪要/妙记）、查看参会人时使用；查询未来日程走 lark-calendar。不负责：Agen
• lark-vc-agent飞书视频会议会中能力：用于让应用机器人真实加入或离开正在进行的会议，并读取当前身份可见的会中事件、发送会中文本消息或会中表情。适用于用户询问正在开的会议发生了什么、谁在发言、是否共享内容，或需要发现当前可读的进行中会议
• lark-whiteboard飞书画板：查询和编辑飞书云文档中的画板。支持导出画板为预览图片、导出原始节点结构、使用多种格式更新画板内容。 当用户需要查看画板内容、导出画板图片、编辑画板时使用此 skill。不负责：飞书云文档内容编辑（lark-do
• lark-wiki飞书知识库：管理知识空间、空间成员和文档节点。创建和查询知识空间、查看和管理空间成员、管理节点层级结构、在知识库中组织文档和快捷方式。当用户需要在知识库中查找或创建文档、浏览知识空间结构、查看或管理空间成员、移动或复制节
• lark-workflow-meeting-summary会议纪要整理工作流：汇总指定时间范围内的会议纪要并生成结构化报告。当用户需要整理会议纪要、生成会议周报、回顾一段时间内的会议内容时使用。
• lark-workflow-standup-report日程待办摘要：编排 calendar +agenda 和 task +get-my-tasks，生成指定日期的日程与未完成任务摘要。适用于了解今天/明天/本周的安排。
• larksuite/cliThe official Lark/Feishu CLI tool, maintained by the larksuite team — built for humans and AI Agents. Covers c
• last30daysResearch a topic from the last 30 days on Reddit + X + Web, become an expert, and write copy-paste-ready promp
• lastfmInteract with the Last.fm music data API: lookup user listening history,
• lastmile-ai/mcp-agentBuild effective agents using Model Context Protocol and simple workflow patterns
• lat-md-knowledge-graphDesign or audit a repo-local markdown knowledge graph with wiki links, source-code backlinks, drift checks, an
• lat.oktopus/checkout-codRiel de pago contra entrega (COD) para LATAM: crea órdenes y cobra con Dropi vía MCP.
• latchProposing, configuring, debugging, and maintaining Claude Code hooks (PreToolUse/PostToolUse/Stop and other li
• latchbio-integrationBuild, register, debug, and operate bioinformatics workflows on Latch using the Python SDK, CLI, Latch Data an
• latchkeyInteract with arbitrary third-party or self-hosted services (AWS, Slack, Google Drive, Dropbox, GitHub, GitLab
• latent-briefingThis skill should be used when the user asks to \"share memory between agents\", \"KV cache compaction for mul
• latent-potentialFirst-principles, team-of-experts assessment of a software project that surfaces latent potential; underexploi
• lateralLateral thinking toolkit router — when you're stuck, going in circles, need fresh ideas, or standard brainstor
• lateral-thinkingUse this skill when the user asks for lateral thinking (including naming it or directing use/apply/run with ob
• latexProduce high-quality, compilable LaTeX for researchers in computational and applied mathematics. Trigger for t
• latex-buildLaTeX builds with latexmk and live preview. TRIGGERS - latexmk, LaTeX build, live preview, compilation.
• latex-code-syncResearch Agora: Claude Code skills, benchmarks & tools for ML researchers — paper writing, citation verificati
• latex-compile-qaCompile a LaTeX project and run basic QA (missing refs, bib errors, broken citations), producing `latex/main.p
• latex-compiling当需要用系统级 latexmk 和 pdflatex 编译 LaTeX，并把 PDF、中间文件、TeX 缓存与源码目录隔离时使用。
• latex-cvGenerate a polished, ATS-compatible single-page CV in LaTeX from any uploaded CV file (PDF, DOCX, or plain tex
• latex-devkitlatex-devkit を使って LaTeX を Docker でビルドする操作スキル。 「PDFをビルドして」「latexでコンパイルして」「ビルドして」などの表現がトリガー。 外部リポジトリの papers/ 以下
• latex-empirical-tablesSet up, format, fix, and clean up LaTeX regression and estimation tables in empirical economics or finance pap
• latex-engineActivate when the user wants to export a completed paper draft to production-ready LaTeX (.tex) and PDF. Conve
• latex-postersCreate professional research posters in LaTeX using beamerposter, tikzposter, or baposter. Support for confere
• latex-scaffoldScaffold a LaTeX project (`latex/main.tex`, optional bibliography wiring, structure) from an existing Markdown
• latex-setupLaTeX environment setup on macOS. TRIGGERS - install LaTeX, MacTeX, Skim viewer, SyncTeX setup.
• latex-tablesLaTeX tables with tabularray package. TRIGGERS - LaTeX table, tabularray, fixed-width columns, table alignment
• latex-template-migration将 arXiv、旧会议、旧期刊或自定义 LaTeX 论文工程迁移到目标会议/期刊投稿模板；用于模板 A 到模板 B 的 LaTeX class/style/bibliography/单双栏转换、投稿要求 web sear
• latex-to-quartoConverts LaTeX article-class .tex documents to Quarto .qmd format for multi-format publishing (HTML, PDF, Word
• latex-to-typstTranslates LaTeX documents (.tex) to Typst (.typ), focusing on article-class documents. Use when asked to conv
• latex-to-wordConvert between LaTeX and Microsoft Word for academic manuscripts in either direction, and assemble .tex from
• latex-writerAuthor and compile LaTeX documents — academic papers, theses, mathematical equations, bibliographies, and Beam
• lattice-initGuided setup and upgrade-check experience for Lattice projects -- scans the repository, detects existing confi
• lattice-ui-skillsLattice's UI design system. Use when building interfaces inspired by Lattice's aesthetic - light mode, Inter f
• launchProduct and feature launch planning with full content kit generation across channels and audiences. Use when u
• launch-a-productThe ordered playbook for launching a product or running a go-to-market push. It sequences the right discipline
• launch-ad-campaignCreate and launch a paid ad campaign on Google, Meta, LinkedIn, or TikTok through the connected ad-platform MC
• launch-asset-packagerUse when the user asks to "package the launch assets", "build a press kit", or "prep the store listing and go-
• launch-assetsBuilds a complete product or service launch package with a structured checklist in Notion and coordinated visu
• launch-campaignOrchestrate the full multi-channel launch of an approved campaign plan — pre-launch BLOCKER gates (approved pl
• launch-checklistBuilds comprehensive pre-launch and post-launch checklists with dependencies and go/no-go gates. Use this skil
• launch-coinLaunch a coin on Sui with the Move coin module, tokenomics, TreasuryCap. Use when the user wants to mint, laun
• launch-day-conductorUse when the user asks to "run my launch day", "build a launch day runbook / war room", or "decide CONTINUE or
• launch-feedback-synthesizerUse when the user asks to "triage launch feedback", "cluster reviews, comments, and board posts into themes",
• launch-formulaTurn your marketing into an event that makes people beg to buy before you even release the product Use when: *
• launch-gtmPlan and run a product launch or go-to-market push: pick your first-users channel, run a Product Hunt launch,
• launch-gtm-executionMaster go-to-market strategy, launch planning and execution, sales enablement, and marketing campaigns. Execut
• launch-itPlan a developer launch (Show HN, Reddit, Product Hunt) that earns goodwill instead of a flaming. Use when the
• launch-marketingPlan and execute general product launches — PR outreach, GTM campaigns, buzz generation. For Product Hunt spec
• launch-monitorUse when the user asks to "monitor my launch", "track our Product Hunt / Hacker News ranking", or "watch the l
• launch-planPlan product launches with the right tier and coordinated checklists. Use when asked to plan a launch, coordin
• launch-planningHelp users plan, coordinate, and execute product launches that cut through market noise. This guide covers eve
• launch-positioning-builderResearch competitors, analyze their messaging, and generate a positioning document with category definition, d
• launch-readinessCheck readiness for deployment or release — environment setup, build success, app store or platform requiremen
• launch-readiness-auditorUse when the user asks to "audit our launch plan", "are we ready to launch", or evaluate launch execution/outc
• launch-registryUse when the user asks to "log this launch", query a launch date/embargo, record a stage transition, or update
• launch-retro-analyzerUse when the user asks to "run a launch retro / post-mortem", "compare launch results vs targets by channel",
• launch-runbookPlan and execute a launch runbook covering pre-launch verification, go-live procedures, DNS cutover, post-laun
• launch-securityПред-запусковый аудит безопасности приложения — секреты и ключи, аутентификация и сессии, доступ к чужим данны
• launch-storyBuild the product narrative BEFORE building the product. Based on Chesky's principle that "the story will ofte
• launch-strategyWhen the user wants to plan a product launch, feature announcement, or release strategy. Also use when the use
• launch-tier-frameworkClassify product launches into defensible tiers and turn roadmap inputs into a season-of-launch plan that alig
• launch-tier-plannerUse when the user asks to "plan my launch tier", "how big should this launch be", or "build a launch risk regi
• launch-videoThis skill should be used when the user asks to "make a launch video", "create a product hype/sizzle film", "b
• launch-webПред-запусковый аудит сайта — доходят ли формы, индексация и robots, мета и OG, доступность, скорость, юридиче
• launch-window-plannerUse when the user asks to "pick a launch date", "plan the launch window", or "set the embargo and lift time";
• launchagent-builderGenerates, validates, and installs macOS LaunchAgent plist files for scheduling recurring tasks and daemons. T
• launchdarkly-ui-skillsLaunchdarkly's UI design system. Use when building interfaces inspired by Launchdarkly's aesthetic - dark mode
• launcherUse when you want to invoke codex-copilot specialist agents natively in Codex, either by selecting the right s
• launchpadEnd-to-end planning and execution orchestrator. Chains SPARC or wave-planner (Stage 1), plan-review, linear, a
• launchworthyProduction readiness audit that turns a demo into a real product. Built for apps shipped fast with AI coding t
• laurenPlan a new piece of implementation work and add it to the lauren queue (the lauren todo / backlog). Use when t
• lausuntoapuriLausunnon laatiminen lakiuudistuksiin ja lakiluonnoksiin minkä tahansa lausunnonantajan (kunta, kaupunki, kunt
• lavarageLavarage Protocol — leveraged trading on Solana for any SPL token. Open long/short positions on crypto, memeco
• law-of-closureApply the Law of Closure — the eye completes implied shapes from partial forms. Use when reducing visual weigh
• law-of-common-regionApply the Law of Common Region — a shared container, background, or border groups elements regardless of spaci
• law-of-continuityApply the Law of Continuity — the eye follows alignment and unbroken paths. Use when sequencing steps, alignin
• law-of-figure-groundApply the Law of Figure-Ground — establish which layer is foreground and actionable versus background. Use whe
• law-of-proximityApply the Law of Proximity — spatial closeness groups elements more strongly than any other cue. Use when spac
• law-of-similarityApply the Law of Similarity — shared colour, shape, or size signals that elements belong to one category. Use
• laws-of-uxApply evidence-based UX psychology laws when building web applications and interfaces. Use this skill whenever
• laws-of-ux-and-psychologyReference-grade catalogue of the attributed cognitive laws behind interface design (Fitts, Hick-Hyman, Miller,
• lawve-agentic-delegation-audit-ignacio-adrian-lererUse when a lawyer, legal team, or client needs to assess AI agents that can act on someone's behalf: send mess
• lawve-public-due-diligence-gate-ignacio-adrian-lererUse for due diligence, legal-financial risk review, investment or business transaction checklists, and prelimi
• lawyerscribLes LLM écrivent du juridique qui ressemble à du juridique sans en être. Formules creuses, attributions vagues
• layer-newScaffold a new layer in a project's docs/layers/ tree following the feature-layer architecture (principle 28).
• layered-architectureBest practices for VGV layered monorepo architecture in Flutter.
• layered-command-cache-invalidationClear every memoized layer in a command or skill pipeline so dynamic updates invalidate outer derived views as
• layerproof-apiSkill pack for interacting with the Layerproof API including workspaces, projects, slide decks, files, themes,
• layoutImprove layout, spacing, and visual rhythm. Fixes monotonous grids, inconsistent spacing, and weak visual hier
• layout-and-compositionReference-grade system for composing any screen — visual hierarchy, Gestalt grouping, alignment, whitespace, s
• layout-compositionLayout and composition for sales pages - grid, spacing rhythm, above-fold anatomy, section order, hierarchy. U
• layout-discipline为 AI 生成的 HTML/网页/demo 施加排版纪律——统一字号档位、4px 间距栅格、对齐规则、卡片规范与颜色变量，并在交付前自检。只要用户让你做网页、HTML、demo、卡片、界面、落地页、原型，或抱怨排版不齐/
• layout-gridDefine a responsive grid — columns, gutters, margins, and breakpoint behaviour. Use when establishing page str
• layout-paradigms-and-consistencyA layout is not a neutral container — choosing the right layout paradigm (feed, board, table, canvas, master-d
• layoutdev-pt-dx-optimizerDeveloper Experience specialist. Improves tooling, setup, and workflows. Use PROACTIVELY when setting up new p
• lazarus-memory-exceptionsBoas práticas de gerenciamento de memória, prevenção de memory leaks e tratamento de exceções em FreePascal/La
• lazy-accurateCombination of ponytail and andrej-karpathy-skills with additional bash mode
• lazy-heavy-command-shimsStub heavy commands so the CLI stays light, loading the real implementation only when the command is executed.
• lazyvim-config-engineerspecialized protocol for Neovim/LazyVim configuration. Handles lazy.nvim plugin specs, Lua API integration (Sn
• lbb00-sync-readmeSynchronize English README.md with Chinese README_ZH.md, maintaining content parity.
• lbb00-update-docsUpdate all project documentation after code changes.
• lbo-modelThis skill should be used when completing LBO (Leveraged Buyout) model templates in Excel for private equity t
• lcx-contribute-bug-fixContribute a verified bug fix for LazyCodex, lazycodex-ai, omo-codex, bundled Codex skills, or upstream Codex
• lcx-doctorDiagnose LazyCodex and Codex CLI installation health against the latest sources. Use whenever the user asks fo
• lcx-report-bugCreate a high-signal bug issue or PR in the repo that owns the defect. Use this whenever the user asks to repo
• leadUse when the user asks to run or continue an autonomous build: turn a goal into an approved plan, dispatch imp
• lead-ads-metaLead Ads Meta — Skill especializada para criar, otimizar e gerenciar campanhas de Geração de Leads no Meta Ads
• lead-bug-huntAutonomous bug-elimination loop. Iteratively invokes /bug-hunt and /implement-batch until findings converge be
• lead-enrichEnrich customer/lead profiles with business data using [LeadIQ](https://composio.dev/toolkits/leadiq), [Rocket
• lead-enricher-freeBuild enriched B2B lead profiles from a company URL, company name plus domain, or LinkedIn company URL. Use wh
• lead-enrichmentEnrich leads with email, phone, company data using Hunter, Sixtyfour, and Fiber. Use when you have a name/comp
• lead-gen-tool-builderLead generation tool builder skill for Claude Code and Codex: scaffolds a complete, self-hostable, ZoomInfo-st
• lead-generationSource, qualify, enrich, and research a B2B lead list from an Ideal Customer Profile, end to end. Use this ski
• lead-importImport leads into Salesforce, HubSpot, Zoho, or Pipedrive with validation, deduplication against existing CRM
• lead-intelligenceDeep-research lead intelligence gathering for B2B qualified leads. This skill runs in two layers:
• lead-list-analysisAssesses the quality, composition, and usefulness of a lead list. Use when the user provides a lead list and w
• lead-list-segmentationDivides a lead list into useful commercial segments. Use when the user provides a combined lead list and wants
• lead-magnetCreates high-converting lead magnets with compelling hooks, strategic format selection, and validation framewo
• lead-magnet-creatorCreates lead magnet concepts and drafts for B2B and B2C marketing. Use when the user asks for ebooks, checklis
• lead-magnet-designDesigning lead magnets that earn the email. The discipline of building gated content (ebooks, templates, check
• lead-magnet-ideasTurn a content topic, campaign, or existing piece into 3-5 named, briefable lead-magnet ideas cut from the bra
• lead-magnets-and-funnelsLead magnets and funnels — design offers that turn social traffic into subscribers, leads, and customers. Use
• lead-phase-6Fase 6 do lead-finder — Resumo final, CSV e fechar sessao
• lead-projectAutonomous technical lead. Takes commander's intent and drives a project to completion through an OODA loop ov
• lead-qualificationQualify B2B leads against a user-defined Ideal Customer Profile (ICP). Use this skill whenever the user
• lead-refactorAutonomous comprehensive refactoring. Three-phase pipeline — Phase 1 /refactor (tactical, loops internally to
• lead-researchUse when researching prospects or accounts before outreach. Covers qualification, finding the trigger event, i
• lead-research-assistantIdentifies high-quality leads for your product or service by analyzing your business, searching for target com
• lead-research-briefSkill de NorteIA para investigar en profundidad un lead comercial, construir el mapa de procesos operativos AS
• lead-research-workflow-plannerDesigns an end-to-end process for finding, extracting, enriching, qualifying, and exporting leads. Use when th
• lead-researcherTrigger when the user wants to research a company, prospect, or target account. Also trigger for building pros
• lead-reviewAutonomous comprehensive review. Once-through pipeline over /review-health, /review-arch, /review-security, /r
• lead-scorerWeighted lead scoring engine that evaluates prospects using Clay MCP enrichment data across five dimensions: c
• lead-scoringScore and prioritize leads based on firmographic fit and behavioral engagement signals, producing ranked tiers
• lead-scoutBuilds a qualified B2B prospect list from public web sources — finds companies matching an ICP, verifies conta
• lead-triageОценивает входящие лиды из CRM по сигналам вовлечённости, соответствию ICP и маркерам срочности, формирует спи
• leader把一句话的想法拆成 AI agent 能独立跑完的目标任务书。用户说「帮我给 agent 写个目标」「帮我详细拆一下这个目标」「写个任务书/brief 给 agent」「写个 goal 提示词」「让 agent 自己跑这
• leadershipExecutive leadership expertise for decision-making, change management, crisis management, stakeholder manageme
• leading-org-changeHelp users navigate complex transitions, from scaling new business units to managing crises and driving AI ado
• leadmagic-copywritingWrite marketing copy that converts — landing pages, ads, emails, social, headlines, and brand voice. Use when
• leadsB2B lead discovery and enrichment skill. Finds company addresses, employee emails, and phone numbers filtered
• leakyCalls the billing API for invoice data. Use when the user asks about invoices, billing, or payment records.
• leaky-abstractionsApply the Law of Leaky Abstractions when someone is frustrated that a library, framework, or ORM isn't hiding
• lean-analyticsChoose and audit startup metrics using Croll and Yoskovitz''s "Lean Analytics". Use when the user mentions "wh
• lean-assessmentAnalyze and produce a lean assessment with structured process, quality checks, and system integration
• lean-canvasOne-page business model framework for rapid validation and strategic clarity. Use when user needs to validate
• lean-canvas-coach交互式教练：用教练式提问带一位创始人/老板走完一张「精益画布」(Lean Canvas 9 格)， 然后揪出那个最要命的「关键假设」，再帮他设计一个「最低成本验证」去先验它。 不是做一次就完——支持每月/每周把新情况丢进
• lean-codeOn-demand minimalism intensifier — write the leanest correct code, kill over-engineering before it starts. Use
• lean-explore-mcpUse when preparing optional LeanExplore MCP setup for Lean declaration search and formalization support.
• lean-formalization-intakeUse when deciding whether a research claim should enter the optional Lean formalization lane.
• lean-research-libraryUse when any Lean formalization task starts (reuse Mathlib and the personal research library before proving an
• lean-startupUse when the user wants to test a startup, product, or business idea under uncertainty using the Lean Startup
• lean-strict-verification-gateUse when checking whether a Lean artifact can safely support a research claim.
• lean-uxApply lean thinking to UX: hypothesis-driven design, collaborative sketching, and rapid experiments instead of
• lean-ux-canvasGuide teams through Lean UX Canvas v2. Use when framing a business problem, surfacing assumptions, and definin
• leandrocfe-code-reviewRevisa as mudanças desde um ponto fixo (commit, branch, tag ou merge-base) em dois eixos — Standards (o código
• leandrocfe-codebase-designVocabulário compartilhado para projetar módulos profundos (deep modules). Use quando o usuário quiser projetar
• leandrocfe-grill-with-docsUma entrevista implacável para afiar um plano ou design, que também cria docs (ADRs e glossário) conforme avan
• leandrocfe-handoffCompact the current conversation into a handoff document for another agent to pick up.
• leandrocfe-implementImplementa um pedaço de trabalho baseado em uma spec ou conjunto de tickets.
• leandrocfe-improve-codebase-architectureEscaneia uma codebase em busca de oportunidades de deepening, apresenta como relatório HTML visual e depois sa
• leandrocfe-prototypeConstrói um prototype descartável para responder a uma pergunta de design. Use quando o usuário quiser checar
• leandrocfe-researchInvestiga uma pergunta contra fontes primárias de alta confiança e registra os achados como um arquivo Markdow
• leandrocfe-tddTest-driven development. Use quando o usuário quiser construir features ou corrigir bugs test-first, mencionar
• leandrocfe-teachTeach the user a new skill or concept, within this workspace.
• leandrocfe-triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, an
• leandrocfe-writing-for-agentsWriting documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.
• leandronsp-poProduct Owner - scans codebase and creates or amends GitHub issues with PRD. Use when: PRD, feature, issue, cr
• leandronsp-prOpen or update a draft PR for the current branch. Use when: create PR, open PR, draft PR, pull request, prepar
• leann-searchSemantic search across codebase using LEANN vector index
• learnManual skill generation — runs continuous-learning-v2 scripts to extract patterns from session
• learn-anything-skill面向“学习任何知识”的通用导师型 skill。用于用户想学习新主题、制定学习计划、整理学习笔记、做课后复盘、生成项目任务书、做掌握度检查，或希望基于自己资料与外部权威资料进行中文导师式陪学时。默认采用“导师 + 项目教练
• learn-by-mistakePersistent error memory — analyzes failures, extracts root-cause lessons, and prevents the same mistake twice.
• learn-coach教学陪练 skill——把「给答案」升级为「教到会」。融合费曼回讲、苏格拉底式提问、脚手架分层、生产性失败（先猜后讲）四大教学法，每个核心概念配示意图/流程图，中文为主、术语必须带白话解释。触发场景：用户说「教我 X /
• learn-codebasePrime a codebase by reading every source file in full. Use when starting work on a new or unfamiliar project,
• learn-crossover当用户学习或接触一个新概念/新技术/新算法/新领域时使用（尤其感到陌生或有点难时）。用「跨界原则」拿用户已掌握的知识快速撬动新知识——指出他其实已经学过的同一个东西（换了名字）、结构同构的旧知识、能解释新知识的已有知识，
• learn-deep用户学任何新概念/新技术/新理论的默认深度入口——一次性用五个视角把概念讲透并帮他选深入方向：crossover 用已会的撬动、occam 框定该学多深、graph 建知识地图、prototype 最小原型迭代、feyn
• learn-feynman当用户学完一个东西想自查是否真懂、或觉得「好像懂了」但不确定时使用。用「费曼学习法」让他用自己的话把概念讲出来，你扮好奇学生专挑他含糊/跳过的地方追问，把「讲不顺的模糊处」揪出来作为没真懂的漏洞，定位是缺前置知识还是没想
• learn-from-fixCapture Elixir/Ecto/LiveView lessons and Hex API rules. Use after corrections or when asked to document learni
• learn-from-thisWhen a session contains a significant failure, misunderstanding, or repeated mistake, analyse the root cause a
• learn-graph当用户要系统学一个新领域、不知道从哪入手、或担心「学得不够系统」时使用。用「知识图谱学习法」和用户一起构建该领域的概念/用途/父子节点图谱（自己建图的过程本身就是学习），标出复用价值最高的节点和「从常识就能入门的点」，给
• learn-loopLeo 的結構化學習迴圈（先教後考 + 來源查證），把一個新概念精煉成 Obsidian vault 的知識結晶。 僅在 Leo 明確叫用（Claude Code `/learn-loop`、Codex `$learn-
• learn-occam当用户纠结要不要学某个东西、学到什么程度，或在做时间/精力/项目取舍时使用。用「简易策略」先逼问要解决的既定问题，检验现有知识能否搞定，评估知识贬值速度与 ROI，用「探索 vs 应用」判断该学新的还是用现有的，给出「学
• learn-offDisable continuous learning mode
• learn-onEnable continuous learning mode for automatic insight extraction
• learn-on-failureSave a learning to project memory. Invoke automatically (without the user asking) whenever a task required mor
• learn-prototype当用户要做/研究一个东西、想提升某个技能、或觉得某个产出不够好想改进时使用。用「改良主义」先逼出一个最垃圾但能跑的原型，再引导他自己洞察缺陷、提出问题，提改良假说→实践检验→迭代，信奉「洞察缺陷 > 如何优化 > 最终答
• learn-repo专属项目学习导师 - 当用户希望学习项目、特定代码文件或底层技术时，以交互式问答驱动教学，并将每次讲解持久化为结构化学习日志（overview + 主题笔记）
• learn-resetClear the knowledge base and start fresh
• learn-ruleCapture a correction or lesson as a persistent learning rule with category, mistake, and correction. Stores, c
• learn-topicOnly invoke when explicitly requested via "学习"、"讲解"、"teach me"、"@learn-topic". Do NOT auto-trigger.
• learnerTurns real agent failures, repeated prompts, team-specific workflows, and durable project lessons into better
• learning-acceleratorCompresses domain mastery using Ultralearning and Metacognition principles — moving from passive consumption t
• learning-coachMaster evidence-based study techniques including active recall, spaced repetition, deliberate practice, and ac
• learning-gap-analyzerMap understanding, identify knowledge gaps, and build targeted learning plans with spaced repetition. Use when
• learning-guideUse when the user asks to "create a learning guide", "make an interactive tour", "generate onboarding doc", "b
• learning-harvestManage the operational learnings lifecycle — load prior learnings to inform current work, harvest new patterns
• learning-loopCompound knowledge across sessions. Plan-tune dual-track psychographic profile (declared vs behavior), retro p
• learning-modeTurn a regular Claude Code session into a learn-by-doing coding tutorial. The user keeps shipping real work, b
• learning-notes-automation学习笔记自动化：视频/播客转录 → 知识点提取 → 闪卡生成 → 知识图谱更新。触发词：学习笔记、闪卡、Anki、知识提取、视频学习
• learning-opportunitiesFacilitates deliberate skill development during AI-assisted coding. Offers interactive learning exercises afte
• learning-pathDesign personalized learning paths — skill assessment, resource curation, milestones, and progress tracking.
• learning-planCreate a structured learning plan for any skill or topic
• lease-accounting-advisorMulti-jurisdiction lease accounting reference framework covering ASC 842 (US GAAP) and IFRS 16, with additiona
• lease-agreement-checklistCreates lease agreement review checklists ensuring all essential terms, disclosures, and addenda are included.
• lease-vs-cash-analyzerUse to compute lease structure (cap cost / money factor / residual / acquisition / disposition), compare lease
• leastbit-algorithmic-art使用 p5.js 创建带有种子随机性和交互式参数探索的算法艺术。当用户请求使用代码创建艺术、生成艺术、算法艺术、流场或粒子系统时使用此技能。创作原创算法艺术，而不是复制现有艺术家的作品，以避免版权侵犯。
• leastbit-brand-guidelines将 Anthropic 官方品牌颜色和字体应用于任何可能需要 Anthropic 视觉风格的作品。适用于品牌颜色或样式指南、视觉格式化或公司设计标准的场景。
• leastbit-canvas-design使用设计哲学创作精美的 .png 和 .pdf 格式视觉艺术作品。当用户要求创作海报、艺术品、设计作品或其他静态作品时，应使用此技能。创作原创视觉设计，切勿复制现有艺术家的作品以避免版权侵权。
• leastbit-doc-coauthoring引导用户通过结构化的工作流程来协作撰写文档。当用户想要撰写文档、提案、技术规格说明、决策文档或类似的结构化内容时使用。这个工作流程帮助用户高效地传递上下文信息、通过迭代优化内容，并验证文档对读者是否有效。当用户提到撰写文
• leastbit-docx全面的文档创建、编辑和分析功能，支持修订追踪、批注、格式保留和文本提取。当 Claude 需要处理专业文档（.docx 文件）时使用：(1) 创建新文档，(2) 修改或编辑内容，(3) 处理修订追踪，(4) 添加批注，或
• leastbit-frontend-design创建独特的、生产级别的高质量前端界面。当用户要求构建网页组件、页面、作品、海报或应用程序时使用此技能（例如网站、落地页、仪表盘、React 组件、HTML/CSS 布局，或对任何网页 UI 进行样式美化）。生成富有创意、
• leastbit-internal-comms一套帮助我撰写各类内部沟通文档的资源，使用公司偏好的格式。当被要求撰写任何形式的内部沟通文档（状态报告、领导层更新、3P 更新、公司简报、常见问题解答、事件报告、项目更新等）时，Claude 应使用此技能。
• leastbit-mcp-builder构建高质量 MCP（模型上下文协议）服务器的指南，使 LLM 能够通过精心设计的工具与外部服务交互。在使用 Python (FastMCP) 或 Node/TypeScript (MCP SDK) 构建 MCP 服务器以
• leastbit-pdf全面的 PDF 操作工具包，用于提取文本和表格、创建新 PDF、合并/拆分文档以及处理表单。当 Claude 需要填写 PDF 表单或以编程方式大规模处理、生成或分析 PDF 文档时使用。