Skills
收录 100,344 个为 Agent 打包的专项能力与操作指引，复制一行命令即可安装。
全部
任何埋点相关工作的强制入口——在读取任何埋点方案文件、查询任何埋点数据或进行任何埋点分析之前，必须首先调用此 Skill。适用于生成、审计、审查、优化、增补、调整、问答、数据验证、故障排查（事件无数据／停止上报／数据异常）以及埋点代码生成——“给我 Android/iOS/Unity 埋点代码”或“埋点代码怎么写”等请求必须先从这里进入；此 Skill 会执行四步识别，然后转至代码生成执行 Skill，因此适用命名预检和方案一致性规则。不要直接根据用户请求调用执行 Skill（代码生成、方案生成、验证），也不要在加载此 Skill 前用 openpyxl/pandas 打开埋点 xlsx 文件或随意翻查日志、文件。它包含规范命名词典、对象分组验证规则、29 个类别的 Markdown 资源（包括部分缓存／共享缓存）、无数据诊断流程，以及用于捕获 TE 平台导入失败的 lint 脚本——这些内容仅在此包中提供，因此随意处理会遗漏阻止导入的问题并走弯路。虽然会多花 1–2 轮，但能防止方案无效和误诊。触发词：埋点、埋点方案、事件设计、埋点审计、埋点评审、埋点优化、增补、递增、SDK 埋点、埋点代码、代码生成、集成代码、数据验证、PRD、故障排查、诊断、无数据、数据缺失、事件未上报。
与没有相关知识的人一起，按照 WCAG 2.2 AA 的要求，以浅显易懂的方式检查并改善小型网站的无障碍功能：对比度、字号和缩放、图像描述、键盘操作、表单、朗读和动态效果。检查报告保存为 docs/barrierefreiheit.md。遇到“Barrierefreiheit”“barrierefrei”“Accessibility”“a11y”“WCAG”“Kontrast”“Alt-Text”“Screenreader”“für ältere Menschen lesbar”“mit Tastatur bedienen”时使用此 Skill。
派遣一个独立于作者的评审者：由全新的评估子 Agent 根据构建前编写的通过标准检查已完成的工作（评估模式），或依据权威数据源审计一批内容（审计模式）；每项结论都需提供证据，且绝不修复问题。所有者要求评估者、审查者、审计员或裁判，或说“派个裁判来”“评估这个”“让一个新的 Agent 检查我的工作”“让第二个 Agent 审查一下”“审计此文件夹”“审计我的注册表”“根据权威数据源验证”或“LLM-as-judge”时使用。仅在被要求时运行，不会在任务结束时自动触发。不用于机器可执行的检查（测试、架构验证、必须返回零结果的搜索）、数百个相同单元的检查，或修复任何内容。
为 Albatros 服务器查找、购买或连接自己的域名（例如 meinladen.ch）。通过地址和电话簿的比喻解释域名和 DNS，提出名称建议，使用 domain_suchen 查询可用域名，并通过 Albatros 和 domain_kaufen 购买（用户是域名持有人，只需在付款页面填写自己的信息）。如果用户已有域名，此技能会使用 domain_verbinden 连接：设置 DNS 记录、TXT 验证 _albatros 并进行检查，直到浏览器中出现带锁的网站。在需要“自己的域名”“我的地址”“购买 .ch”“哪些域名可用”“搜索域名”“DNS”“A 记录”“CNAME”“TXT 记录”“域名没有指向我的网站”“证书”“https 无法使用”时使用此技能。
通过船上的航海日志和港口副本（Albatros 的 Git 服务 code.aufwind.sh）的比喻，为没有相关经验的人讲解 Git。指导用户从港口获取仓库，通过优质备注完成首次记录（Commit），将记录发送到港口（Push），并检查副本是否到达 Albatros 服务器的 ~/projekt。规则：仅使用 main，只逐个暂存自己的文件，绝不 force-push。在需要“Git”“GitHub”“港口”“commit”“push”“pull”“Repo”“什么是 Repository”“保存我的更改”“push 被拒绝”“冲突”，以及替用户提交任何内容之前使用此技能。
规划并陪伴项目从 Albatros 迁移到自己的服务器，并用搬进自家房子的比喻来讲解：迁移清单、租用并配置服务器（SSH 仅使用密钥、配置防火墙和更新、用 Caddy 配置 HTTPS）、将 GitHub 上的 Logbuch 一并迁移、通过 SSH 使用自己的密钥将那里的 Fliessband 切换到新目标、迁移数据和机密信息、测试，以及更改 DNS 地址，或使用 domain_mitnehmen 中的授权码将从 Albatros 购买的域名转移到新注册商。备份由自己负责。在需要“自己的服务器”“迁移”“离开 Albatros”“Hetzner”“VPS”“自行托管”“落地”“拥有自己的服务器需要什么”时使用此技能。
Albatros 的主 Skill：空中交通管制员。使用工具 lage 查询用户的实际进度（订阅、密钥、服务器或网站、流水线、域名、任务），然后准确调用当前所需的 Skill；每完成一步后都会重新询问。未登录时，会显示所有 Skill 的开放库供用户自行选择，并说明如何登录。只要有人继续处理 Albatros 项目，或询问“现在怎么办？”“接下来怎么做？”“下一步是什么？”“我进展到哪了？”“不行”“卡住了”“我不知道该怎么办”，或使用 /albatros:lotse，以及在 start Skill 结束时，都使用此 Skill。对于明确的主题问题（例如“如何连接我的域名？”“什么是 SSH 密钥？”），应优先使用对应的专业 Skill。
Use when the user asks to brainstorm, think an idea through, or explore what to build, or brings a rough idea whose goal or reason is not stated yet ("add notifications", "something for onboarding", "I have an idea for an app"), or asks for ideas or options ("give me ideas for X", "what could we do about Y"), in software or outside it. Works in Claude Code and Codex. Produces a design brief the user approved, ready for mstack-write-specs.
Use when the user wants Codex to write the code while Claude manages, reviews and commits it, for a plan or plan folder, a list of issues (GitHub or another tracker) or one larger task, such as "let Codex develop this plan", "Codex writes, you review", "use the Codex plugin to implement these issues", or when an mstack-plan execute-plan.md offers this workflow. Runs in Claude Code only, with the OpenAI Codex plugin; for Claude subagents as writers use mstack-codex-review-workflow. Produces one integration branch with one reviewed commit per phase.
Use when review feedback has arrived and has to be acted on, such as findings from a cross-vendor reviewer, the advisor, a fresh-context reviewer, CI review bots or comments on a pull request, or when the user says "process this review", "address the review comments", "handle these findings" or pastes reviewer output, or when another mstack skill needs a review processed. Works in Claude Code and Codex. Produces verified fixes with tests, reasoned rejections, and a log with a drafted reply per finding.
用带质量检查的流水线作比喻，向没有相关知识的人解释并设置 CI/CD：Albatros 港口（Forgejo，code.aufwind.sh）中的 Workflow 会在每次推送到 main 时启动。自有服务器上的 Worker（Runner）会构建网站，通过健康检查后发布，并通过公网地址检查；否则会回退到上一个正常版本。遇到“Pipeline”“CI/CD”“GitHub Actions”“Forgejo Actions”“Workflow”“Runner”“automatisch ausliefern”“deployen”“wie kommt meine Änderung auf den Server”“Fliessband ist rot”“Deploy fehlgeschlagen”时使用此 Skill。
Find out whether ChatGPT, Claude, Perplexity, Google, Bing/Copilot, Apple, Meta and other AI crawlers can actually reach your site, and fix what blocks them without opening up what should stay private. Use when the user says or asks "is my site blocking AI crawlers", "can ChatGPT see my website", "check my robots.txt for AI bots", "should I block GPTBot", "GPTBot vs OAI-SearchBot", "block AI training but stay in ChatGPT search", "allow ClaudeBot", "what is Google-Extended", "why doesn't Perplexity cite us", "Cloudflare is blocking AI bots", "are AI bots getting 403s", "opt out of AI training", "noai meta tag", "X-Robots-Tag for AI", "fix robots.txt for AI search". Runs a dependency-free Python checker against the live site and reports each crawler by name and job. Works without a Revnu account.
Audit whether your website gives ChatGPT, Perplexity, Claude, Gemini and Google's AI Overviews something worth quoting, and fix the pages that don't. Use when the user says or asks "is my site ready for AI search", "audit my site for AI answers", "AEO audit", "GEO audit", "why doesn't ChatGPT quote our pricing", "make our pages citable by AI", "do I need llms.txt", "does schema help with ChatGPT", "check our structured data", "is our content visible without JavaScript", "AI SEO audit", "optimize our homepage for AI answers", "LLM SEO check", "why does AI get our product wrong", "fix our site for AI Overviews". Checks fetchability and indexability, answers in the raw HTML, entity facts, schema (including JS-injected markup), headings, dates and llms.txt, then gives at most 10 prioritised fixes and edits the repo when asked. Works without a Revnu account.
Fix what ChatGPT, Perplexity, Google AI Overviews, Copilot or Claude get wrong about your business: list every claim they make, mark it correct, wrong, outdated or missing, find the page it most likely came from and draft the fix (your own copy, your profiles, a correction request). Use when the user says "ChatGPT says the wrong price", "AI thinks we're a different company", "Perplexity says we don't have a free plan", "AI answers describe our old product", "Google's AI Overview is wrong about us", "ChatGPT confuses us with another company", "fix what AI says about us", "AI says we lack a feature we have", "prospects quote wrong things from ChatGPT", "correct our brand in AI answers", "why does Copilot say that about us", "AI uses our old name", "update AI after our rebrand or pricing change", or hands over a revnu-geo-check report. Works without a Revnu account.
Find out whether AI answers mention and cite your business when buyers ask about your category, what they say about you, and which pages they cite instead. Use when the user says or asks "does ChatGPT recommend us", "are we showing up in AI search", "check our AI visibility", "GEO check", "AEO check", "what does Perplexity say about us", "do AI answers cite our site", "why does ChatGPT recommend our competitor", "AI share of voice", "test prompts in ChatGPT", "track our brand in AI answers", "is Gemini getting our pricing wrong", "which sites do AI answers cite for our category", "LLM visibility report", "how often are we mentioned by AI". Runs buyer prompts several times with control prompts on the surfaces this host can actually reach (its web search tool; OpenAI, Perplexity or Gemini APIs only if keys are already set and the user approves the cost), reports rates with sample sizes, and lists every claim made about the business with the page to fix. Works without a Revnu account.
Get onto the third-party pages AI answers and search cite when buyers ask for the best tool in your category: "best X" roundups, directories, review sites, GitHub awesome lists, community threads and comparison pages. Finds which ones are cited, ranks them by fit and effort, and drafts each submission, PR or pitch by the target's own rules. Use when the user says "get us listed", "which directories should we submit to", "how do we get into best-of lists", "why does ChatGPT recommend competitors and not us", "get mentioned in AI answers", "submit to awesome lists", "get on G2", "where are our competitors listed", "pitch listicle authors", "find backlink opportunities", "launch directories", "get into roundups", "Product Hunt alternatives", "where should we be listed", "get cited by Perplexity". Never submits, pays or posts without a yes. Works without a Revnu account.
Write down once who buys, what they're stuck on, what it costs, what proof you have and what you've already tried, so every growth task starts from facts instead of a questionnaire. Creates or updates .agents/revnu-growth.md from the repo, the live site and the user, and imports an existing product-marketing.md. Use when the user says "set up our marketing context", "who is our ideal customer", "write our ICP", "describe the product for marketing", "positioning doc", "what's our brand voice", "remember this about our customers", "log what we tried", "what have we learned from marketing", "we changed pricing, update the context", "update our growth notes", "summarize the business for the agent", "import my product-marketing.md", "start a growth doc", or before a first marketing, SEO, outreach or launch task when no context file exists. Works without a Revnu account.
为没有相关知识的人介绍其 Albatros 服务器。用住宅中的事物作比喻，解释终端、文件夹、存储空间、内存、运行中的程序和主钥匙 sudo。用户亲自输入第一批命令。遇到“was ist auf meinem Server”“wie bewege ich mich im Terminal”“was ist ein Ordner auf dem Server”“ls cd pwd”“wie viel Platz habe ich”“was läuft da”时，以及首次成功通过 SSH 访问后，使用此 Skill。
一个开放协议，规定了 AI 应用如何以统一方式连接外部工具与数据源。客户端负责与模型对话，MCP 服务负责提供工具、资源和提示模板，两者通过标准消息通信，同一个服务可以被不同客户端复用。