Product Development
gustavscirulis/snapgrid
End-to-end product development for iOS/macOS apps. An agent skill from gustavscirulis/snapgrid.
独立开发者的项目全生命周期管理:需求验证、选词选品、建站或做原生 App、上线接入、SEO/GEO 获客、支付变现、监控迭代与跨会话接力。以下情况使用:用户提到 rankup 或 /rankup(check、review、init;doctor 仅指整理项目 .rankup 记录);当前目录或工作区有…
$ npx skills add yan-labs/yan-skills --skill rankup -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install yan-labs/yan-skills rankup --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/yan-labs/yan-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/rankup .claude/skills/rankup && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "rankup" agent skill from https://github.com/yan-labs/yan-skills/tree/main/rankup into .claude/skills/rankup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "rankup", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/yan-labs/yan-skills/tree/main/rankupType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add yan-labs/yan-skills --skill rankup -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install yan-labs/yan-skills rankup --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/yan-labs/yan-skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/rankup .agents/skills/rankup && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "rankup" agent skill from https://github.com/yan-labs/yan-skills/tree/main/rankup into .agents/skills/rankup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "rankup", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add yan-labs/yan-skills --skill rankup -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install yan-labs/yan-skills rankup --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/yan-labs/yan-skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/rankup .cursor/skills/rankup && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "rankup" agent skill from https://github.com/yan-labs/yan-skills/tree/main/rankup into .cursor/skills/rankup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "rankup", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/yan-labs/yan-skills.git --path rankup--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add yan-labs/yan-skills --skill rankup -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install yan-labs/yan-skills rankup --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/yan-labs/yan-skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/rankup .gemini/skills/rankup && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "rankup" agent skill from https://github.com/yan-labs/yan-skills/tree/main/rankup into .gemini/skills/rankup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "rankup", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install yan-labs/yan-skills rankupInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add yan-labs/yan-skills --skill rankup -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/yan-labs/yan-skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/rankup .github/skills/rankup && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "rankup" agent skill from https://github.com/yan-labs/yan-skills/tree/main/rankup into .github/skills/rankup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "rankup", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add yan-labs/yan-skills --skill rankup -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install yan-labs/yan-skills rankup --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/yan-labs/yan-skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/rankup .opencode/skills/rankup && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "rankup" agent skill from https://github.com/yan-labs/yan-skills/tree/main/rankup into .opencode/skills/rankup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "rankup", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
rankup独立开发者的项目全生命周期管理:需求验证、选词选品、建站或做原生 App、上线接入、SEO/GEO 获客、支付变现、监控迭代与跨会话接力。以下情况使用:用户提到 rankup 或 /rankup(check、review、init;doctor 仅指整理项目 .rankup 记录);当前目录或工作区有…
Rankup is an agent skill from yan-labs/yan-skills. 独立开发者的项目全生命周期管理:需求验证、选词选品、建站或做原生 App、上线接入、SEO/GEO 获客、支付变现、监控迭代与跨会话接力。以下情况使用:用户提到 rankup 或 /rankup(check、review、init;doctor 仅指整理项目 .rankup 记录);当前目录或工作区有 .rankup/,或要读取、续做、整理项目计划、路线图、待办、PRD、交接或进度文档;在这些项目记录上下文里说「继续」「接着做」「上次做到哪」「下一步做什么」;以及项目规划、需求验证、关键词调研与 SERP、独立 Google Trends 热度查询或对比、AI 搜索推荐(GEO)、网站体检、sitemap/IndexNow/Search Console、流量、建站、上线、支付(Stripe、Anyway、PayPal)、变现与增长。SEO/GEO 是主要手段,不是适用边界,非 SEO 的项目计划与 macOS、iOS、iPad 原生 App 同样适用。纯文案(含 SEO 趋势博客等主题写作)、纯视觉设计、与项目管理无关的通用开发及基础设施排错(含 Cloudflare 部署报错、普通 CLI…
Its SKILL.md is about 5.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 269 other files, including scripts and reference files (for example `data/provider-capabilities.json`, `data/word-roots.json` and `evals/evals.json`).
It sits in Marketing & SEO, covering Link building, Image generation and PRD writing. It works with Cloudflare, iOS, macOS and Google Search Console. The repository describes itself as: Yan's agent skills collection — Google Trends SEO workflows, AI news, autopilot, and more. For Claude Code / Codex / Cursor. The licence is MIT.
3 steps, taken from the first numbered list in SKILL.md.
Read from SKILL.md and the folder at commit 8e35596. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Ships 1 file in scripts/, which the agent can run.
Shell commands in SKILL.md call:
npxnodegitnpmFrom the folder's file list and the shell code blocks in SKILL.md.
Links to these hosts (documentation or services it may open):
seo.web.cafeFrom URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Rankup loads about 5.1k tokens when it runs, and up to ~418k if it reads all its reference files. Until then it costs about 153 tokens; SKILL.md has 987 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check noted patterns worth knowing about, such as sudo or a known installer.
- 第三方工具令牌只有一份,放 Skill 根目录 `.env`,环境变量优先;细则见 `discipline.md` 十一。Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
The full file from yan-labs/yan-skills at commit 8e35596, republished under its MIT licence (© yan-labs). 987 words, ~5,071 tokens.
.claude/skills/rankup/SKILL.md (or your agent's skills folder). This skill also uses 264 other files; get the full folder from GitHub.Rankup 是独立开发者的项目全生命周期管理 Skill:调研与需求验证 → 立项与定位 → 建站或做原生 App(macOS、iOS、iPad;不做 Android App,Android 商店只作需求参考)→ 上线前 SEO / GEO → 上线与接入 → 外链获客 → 变现与监控迭代。
SEO 与 GEO 是最主要的获客与验证手段,不是适用边界:项目计划、交接文档、功能迭代、上线运维、项目记录整理都归这里。
每个项目的 .rankup/ 记录长远规划与当前进展,负责跨会话接力;本文件只做入口与路由,细节在 references/,一次任务只读需要的那部分。
取不到的数据写「未知 / 不可用」,不写 0;区分「官方要求」「工具建议」「项目经验」;阶段性汇报不等于完成。
项目任务按下表读取相关入口与接力,再选择任务分支,避免依据过期印象执行;纯只读与轻量任务按对应例外执行。
| 触发条件 | 必须先读 | 必须做 | 完成标志 |
|---|---|---|---|
| 开工:项目任务(含工作区根) | 接力协议、项目接力记录 | 按接力与下文启动协议恢复、对账和记录状态 | 接力可续做 |
| 里程碑:产物、决定或后台任务状态变化 | 同上接力协议 | 覆盖当前接力与产物指针 | 接力反映最新状态 |
| 收尾:完成、放弃、叫停或等用户决定 | 维护「一」「二」 | 按轻量路径或五步维护收口;扫描用 scripts/maintain/ref-scan.mjs | 回复有维护摘要与证据 |
| 有项目记录上下文的「继续」「接着做」「上次做到哪」「我们开始执行这个项目的计划」,或读计划 / 交接 / 进度文档 | 同上「新会话怎么接」 | 定位项目 → 读接力(无 .rankup/ 用 HANDOFF.md)→ 核对后台产物与改动 → 直接执行下一步 | 下一步已有执行证据 |
| 选词、调研、找需求(已有主词) | playbooks/entry.md → research.md | 按入口四步顺序与两路判读执行;方向未落主词先走 selection.md,无方向先走 P1,主词出现后回本行 | 入口卡写进 .rankup/research/<词根>-<日期>.md |
| ChatGPT 问法链路、自然推荐验证 | seo-geo.md 步骤 2–3 | 问法、采样、追问与三清单只按该唯一源;词级编排与裁决回 entry.md | 入口卡②有三清单与原始证据指针 |
| 「谷歌质量」「HCU」「E-E-A-T」「AI 内容会不会被打」「内容 review」「垃圾内容政策」 | seo-helpful-content.md | 区分官方口径与业内推断;按单页或站点级自检表核对并留证据 | 对应自检表通过且无红线 |
| 建站、做功能、写 UI | lifecycle/stage-3-build.md、cloudflare-stack.md | 用 shadcn 脚手架与组件库,不手写基础控件;当天过完 Day-1 清单 | checklists.md 段 3 通过 |
| 做内页、「把这个词做成页面」 | lifecycle/stage-4-prelaunch.md「新增内页随手清单」 | 页面映射按 段 2 §2.3;目标词登记到 keywords.md;按段 4 完成页面与体检 | checks.md 记 ✅ 与证据 |
| 上线前体检、「能不能上线」 | checklists.md 段 4、lifecycle/stage-4-prelaunch.md | 全套体检:AITDK、PageSpeed、图标专项、占位专项;重跑不采信上一轮 | 段 4 闸门全绿,或写明改不动的理由 |
| 接入、绑域名、提交 sitemap、「怎么不收录」 | lifecycle/stage-5-launch.md、search-platforms.md | 先 site-onboard.mjs --check 再执行;接入一律线上实测 | .rankup/integrations.md 逐行有证据与日期 |
| 外链 | backlink Skill、lifecycle/stage-6-backlinks.md | Rankup 只判时机与数量;提交前把候选清单给用户圈定 | 台账有 submitted → public → indexed 证据 |
| 变现、支付、定价 | monetization.md、lifecycle/stage-7-monetize.md | 按分发方式选主通道与备份,目标环境端到端验证 | 验证记录进 .rankup/integrations.md |
| 维护 Skill 本身:改 rankup 源码,或「把这个经验写进 rankup」 | maintenance.md「五」、evolution.md | 全量扫描后改;拆移文档用 scripts/maintain/split-doc.mjs;scripts/maintain/doc-lint.mjs 断链为 0;validate 与相关测试通过;按规则升版本 | node scripts/validate-rankup.mjs 通过 |
「/rankup doctor」「doctor」(整理项目 .rankup/ 记录)、「整理一下项目记录」「.rankup 太乱了」「整理一下 rankup」「把这个项目的 .rankup 重新整理」 | maintenance.md「六」、project-memory.md 目录规范 | 免版本更新与线上对账;rankup-doctor.mjs 只读诊断 → 经验分拣与回流(按四层归属逐条判去向)→ 分三类出计划 → 执行 A、B → 回读校验;C 类与别的 Skill 的经验只列给用户 | 整理报告:改了、删了(从哪找回)、留了、经验分拣表、待用户决定 |
本表是已进入 rankup 项目流程后的路由,不扩大触发范围。
用户不会说「跑一下 seo-audit.mjs」。命中就照入口走,不要现编步骤;越模糊越不盲跑全套:有明确对象先选对应 playbook,已有项目没有明确动作先 rankup check,连项目与方向都没有时只问一句有没有想好的词或方向。
| 用户会说的话 | 段 | 入口 |
|---|---|---|
| 「做网站时需要 AI 视频/图片生成」「接 Kie」「Kie 音频可不可以用」 | 3 | Kie 可选分支 → 模型快照、登记,按能力选供应商,Kie 不是必选 |
| 有项目记录上下文的「继续」「接着上次做」「上次做到哪了」「看看这份计划」「我们开始执行这个项目的计划」 | 当前段 | 见「强制流程」开工与继续行 |
| 「挖点需求」「最近有什么能做的」「找几个关键词」(没给具体词) | 1 | research/p1-discovery.md |
| 「调研一下这个词」「这个词能不能做」「帮我扩词」「找个 xxx 的需求」 | 1 | entry.md → research/p2-keyword-root.md(任何词都是词根,按五个取数动作全自动跑完,不反问) |
| 「做小语种」「这个词在德语怎么搜」「某国市场找词」 | 1 | P2 阶段 0.7 开工卡 + 三关;语种探测 trends.md W1 |
| 「找个方向」「这个方向值不值得做」「选品」「帮我看看这个想法」 | 1 | selection.md 收集方向候选;主词出现后按 entry.md 验证与裁决 |
| 「App 有没有需求」「找 iOS/iPad/macOS 产品」 | 1–2 | research.md App 分支 → 段 2 2.2 |
| 「谁在赚钱」「反查这个站」「帖子说月入 X 是真的吗」 | 1 | research.md P4 + demand-sources/validation-chain.md |
| 「筛这批 AITDK 报告」「只看竞品异常」 | 1 | seo-box.md:aitdk-triage.mjs 离线分流 |
| 「运行趋势监控」「扫描种子 Rising」「美日韩趋势追踪」 | 1–2 | trends-system.md:三区串行扫描、去重、健康台账与追踪复查 |
| 「XX 和 YY 哪个更火」「今天美国在搜什么」「这个词有没有量」 | 1–2 | trends.md(gpts 基线判读),scripts/gt.py |
| 「这个词在 AI 里有多大需求」「候选会不会被 ChatGPT 推荐」「竞品为什么被 AI 推荐」 | 1 / 4 验证 | 词级编排先读 entry.md,问法链路只按 seo-geo.md;面板数据见 provider-capabilities.md「三·五」 |
| 「已上线站 ChatGPT 会不会推荐我们」「问 ChatGPT 为什么不推荐」「推荐位整改」「GEO 闭环」「复审」 | 5 / 6 上线后 | seo-geo-recommendation-loop.md:复用问法与三清单 → 建议甄别 → 整改发布 → 复审;采样规则只指 seo-geo.md |
| 「有什么游戏站能做」「小游戏机会采集 / 决策」 | 1–2 | game-sites.md:平台监控、sitemap 差分、新标题筛选;旧每日决策协议仅部分覆盖 |
| 「做不做」「做哪个语种」「要不要多语言」「做成工具还是内容站」 | 2 | lifecycle/stage-2-positioning.md |
| 「我们做个网站吧」「帮我搭起来」 | 2→3 | 先过段 2 立项,再段 3 初始化;手上没有词树先回段 1 |
| 「做个功能」「加个 X」 / 开发中「按 rankup 规范来」 | 3–4 | lifecycle/stage-3-build.md:UI 一律来自 shadcn 组件库,多功能工具站导航先读侧栏统一规范;做完段 4 全套体检 |
| 「做个好看的页面」「Hero 怎么设计」「找个组件参考」 / 任何整站设计、改版 | 3 | 必须走 /design 并逐字带上获奖标准 prompt(全文见 stage-3-build.md 硬规则「设计一律走 /design」),再用 design-references.md 选 2–3 个案例再实现 |
| 「我们做个内页吧」「关键词没问题了,做成内页」 | 4 | 见「强制流程」做内页一行 |
| 「看一下 SEO / GEO 有没有问题」「能不能上线」「TDK」「密度」 | 4 | checklists.md 段 4 + seo-box.md;第三方复核加载官方 gefei-page(seo-webcafe.md) |
| 「怎么被 AI 引用」「llms.txt」「对 AI 代理友好吗」 | 4 | seo-ai-search.md;推荐位做法见 seo-geo.md;SEO 专题索引 seo-growth.md |
| 「生成 logo / 配图 / og 图」「做一整套网站图标」 | 3–4 | /imagegen 真实生成;图标集 scripts/make-favicons.mjs --src <logo.png> --out <public目录> |
| 「写一下这页的文案」「AI 味太重」 | 4 | 中文 /write(附属 Skill 缺的用 find-skills 装齐);内容形状按 /ai-seo,见 skill-ecosystem.md |
| 「站慢不慢」「Core Web Vitals」 | 4 | scripts/pagespeed.mjs collect --strategy both,判读 seo-box.md 一 |
| 「这个域名能不能用」「域名黑历史」 | 5 | 段 5 黑历史闸门 + 官方 gefei-domain |
| 「域名买完了」「帮我绑域名」 | 5 | cloudflare/domain-email.md §8.5 全 API 绑定 → cloudflare-stack.md §8.8 基础安全 → domain-email.md §8.6 邮箱 → 上线验收 |
| 「数据平台都接了吗」「GSC 接了没」「提交 sitemap」「怎么一直不收录」 | 5 | node scripts/site-onboard.mjs --domain <域名> --repo <仓库> --check;search-platforms.md、analytics-platforms.md |
| 「把 Ahrefs 的检验结果都修了」 | 5→4 | scripts/ahrefs-site-audit.mjs 取清单 → scripts/ahrefs-issues-recheck.mjs 线上复核 → 修完段 4 全套重跑 |
| 「帮我搞点外链」「竞品外链哪来的」「发个 Product Hunt」 | 6 | backlink Skill + webcafe-topics.md 五;发布平台 product-launch.md |
| 「没人付费」「文案怎么写才有人点」「定价页怎么排」「渠道」「定价怎么定」「接 PayPal / Anyway」「AdSense 被拒」 | 7 | monetization.md、payments-anyway.md、conversion.md;角度用 /marketing-psychology、/marketing-ideas |
| 「流量掉了」「排名没了」「是不是被 K 了」 | 7 | webcafe-experiences-2.md 十七~十九 |
| 「流量涨了」「这两天为什么涨」「why is traffic up」 | 7 | 先读项目 .rankup/INDEX.md 与 .rankup/decisions.md,结合线上页面和代码核对已下线功能、重定向及实际能力;再按 GSC 日期/页面/查询与分析来源拆增量,URL 迁移按 seo-experiences-2026-07-late.md「[2026-07-30] URL 迁移期的页面报表必须把旧 URL 与新 canonical 合并看」判读。未显示查询与未定稿日期单列未知 |
| 「现在该做什么」「一步步来」「到哪一步了」「让流量涨一点」「优化一下我的网站」「今天弄下 SEO」 | check | rankup check(下文「命令」);默认打磨转化链路,见段 7 段首 |
| 「review 一下我的站」「查漏补缺」「这项目脱轨了」 | review | rankup review |
| 「把这个老项目接进来」「rankup init」 | init | rankup init → project-memory.md |
「/rankup doctor」「doctor」(整理项目 .rankup/ 记录)、「整理一下项目记录」「.rankup 太乱了」「整理一下 rankup」 | doctor | rankup doctor → maintenance.md「六」 |
| 「群里怎么说的」「哥飞说过什么」 | 经验 | 官方 gefei Skill 知识库工作流;公开论坛 webcafe-forum.md |
| 「把这个经验写进 rankup」 | 维护 | maintenance.md「五」第 2 条 |
| 「抓一下后台数据」「Semrush 能查这个吗」 | 取数 | 哥飞工具先加载官方 gefei;其他面板看 provider-capabilities.md |
| 说的事这张表没有,或点名的兄弟 Skill 本机没装 | — | capability-map.md → skill-ecosystem.md → integrations.md 用 find-skills 搜索并安装;不现写等价实现 |
每段保留适用场景、入口、原则摘要与闸门指针。硬规则全文(含「为什么」)在各段文件,步骤 check 与操作也在那里;lifecycle.md 是总述、旧编号映射与对账。完成验收只在 checklists.md,关键词与立项裁决只在 entry.md。
selection.md)。research.md(P0 分流:没东西 P1、一个词 P2、一个域名 P4);裁决只见 entry.md,历史经验见 demand-discovery.md;证据覆盖见 research-checklist.md。常用:scripts/demand/suggest.mjs、官方 gefei-keywords / gefei-competitor、backlink/scripts/semrush-keyword.mjs、面板取证 scripts/rankup-cli.mjs(npx @yan-labs/rankup audit similarweb);问法与采样工具见 seo-geo.md。seo-geo.md。checklists.md 段 1。entry.md。立项后的经营经验见 zero-to-one.md。entry.md;市场选择、产品形态及问法簇到页面的移交见 段 2。checklists.md 段 2。build-macos-apps 专项 Skill,iOS/iPad 按对应原生工具,商店分发读 monetization.md 五;只有 App 时网页项标 N/A。cloudflare-stack.md(脚手架命令、资源选择);三方库优先见 integrations.md。checklists.md 段 3。seo-box.md、seo-webcafe.md、seo-ai-search.md。常用:scripts/seo-audit.mjs --sitemap、scripts/pagespeed.mjs collect --strategy both、scripts/is-agentic.mjs scan --save、scripts/ai-crawler-access.mjs --url <正式域名>、scripts/ua-parity.mjs <url>。checklists.md D1 / D4 / D12 / D13 / P3(索引水合、Schema 语义、网格与键盘、SSR 可达性、分析去重与真实上报);操作见段 4 文件与 analytics-platforms.md。checklists.md 段 4。search-platforms.md、analytics-platforms.md、域名接入 cloudflare/domain-email.md。常用:scripts/site-onboard.mjs、scripts/gsc-domain-verify.mjs、scripts/bing-import-from-gsc.mjs、scripts/cf-analytics-setup.mjs、scripts/indexnow-submit.mjs、scripts/webmaster-sitemap.mjs、scripts/yandex-setup.mjs、scripts/ahrefs-site-audit.mjs、scripts/analytics-beacon-check.mjs。checklists.md 段 5。backlink Skill(未装:npx skills add yan-labs/yan-skills --skill backlink -g -y);判据 webcafe-topics.md 五;发布平台 product-launch.md。checklists.md 段 6。monetization.md、conversion.md、evolution.md;掉量排查 webcafe-experiences-2.md 十七~十九;常用 scripts/is-agentic.mjs diff、scripts/review.mjs。checklists.md 段 7。先查脚本清单,禁止现写等价实现或手点界面,脚本坏了修脚本(discipline.md 二);执行、取数、完成标准与事故判据统一查 discipline.md,AI 探针的通道与证据边界查 seo-geo.md。
checklists.md 段 5。每段都有一套 checklist,不过 check 不许进下一段;每轮迭代新做的东西,把相关 check 重新过一遍。
checklists.md,判「这段能不能算完」;步骤 check 在各段文件,判「这一步做对了没有」。.rankup/checks.md:✅ + 证据位置 + 日期;做不了标 ⏸ 写清卡在哪;开新一轮把标「每轮」的打回 ⬜。scripts/review.mjs 只给文件层面的缺口。checklists.md,操作只在各自文档;缺 check 先补进 checklists.md 再去做。rankup check「现在该做什么」「一步步来」时的唯一动作,编排在 playbooks/site-review.md 第二节。读 checklists.md 与 .rankup/checks.md,跑一次 scripts/review.mjs,找到第一个没过闸的段,逐项去真实代码、线上响应、后台读数核对,直接照着做并在 checks.md 记证据。零配额、不派七组 agent;命中升级判据(已上线但 audit.md 缺失、.rankup/ 不存在、动过线上 URL 且超过一轮没体检、用户问的其实是站有什么问题)就明说「这已经不是 check,是 review」并转全站体检。
rankup init全新项目,或做了很久还没有 .rankup/ 的项目(常态,不得因为缺记忆就重建技术栈):摸清 package.json、路由、部署配置、git log,已上线的再取线上 sitemap.xml、robots.txt、首页;外部系统一律实时查询;按 project-memory.md 建 .rankup/ 全套,取不到的写 待确认;integrations.md 用完整平台表初始化全部 ⬜(discipline.md 十);已运行项目补 baseline.md 与 audit.md,roadmap.md 写阶段目标与放弃条件;绿地项目脚手架跑通即建私有远端仓。已有 .rankup/ 时不覆盖,转为补齐并提示用 review 或 doctor。
rankup review对站本身做全面体检,编排在 site-review.md 第一节:先摸前提,再把可独立运行的组按可用并发派出(技术 SEO / 速度 / GEO / 关键词长尾 SERP / 哥飞官方 Skill 数据复核 / 市场规模 / 接入与记忆),主 Agent 汇总回写。A 组含 AITDK 全站报告,Issues 与未满分项一律必修(checklists.md「闸门 4c」)。G 组:scripts/review.mjs --project-root . 出五块报告;会话记录 scripts/sessions.mjs --project-root . --days 14 --new-only(水位线在 .rankup/review-state.json,--dump 消化完才 --mark),从中找用户纠正、验证过的结论、坑与根因、已推翻旧记录的事实(修订原条目)。review 发现 Day-1 项缺失,先回流进段 3 / 段 4 清单再修站。
rankup doctor整理项目记录,是「维护」章节的显式入口:在用了很久、文档很多的项目上,按 maintenance.md「六」 全量扫描、做经验分拣与回流(项目 / 用户全局层 / Skill 源码 / 别的 Skill,判定见「七」)、清理过时 / 重复 / 任务型内容、原地修订被推翻的结论、目录归位、校验根层 portfolio.md 与 projects.json 条目数、输出整理报告。第一步永远是只读诊断 node "<rankup-skill-dir>/scripts/maintain/rankup-doctor.mjs" --project-root . [--portfolio-root <工作区根>];拿不准的只列给用户,不删。与 check(定位下一步)、review(修站)的分工见该节表格。
skill.json,跑 node "<rankup-skill-dir>/scripts/check-version.mjs" --project-root . --apply;网络失败保留当前版本,不得伪称已更新。doctor 只整理记录,免版本更新与线上三方对账;任务明确禁止联网、Git 或改动时遵守其范围。git log --oneline -25、真实路由与线上 sitemap.xml 全量 <loc>;按任务适用面核对,记录与现实不一致先回写。只读问答只核对所问事实,不为此制造状态变化。project-memory.md。references/experiences/ 回答「该怎么判断、别人踩过什么坑」:挖需求读 demand-discovery.md,规划读 zero-to-one.md,上线后改什么读 conversion.md,技术 SEO / 站群 / 多语言 / 索引读 webcafe-experiences.md 与 webcafe-topics.md。经验层不带项目信息;每条有出处与证据等级(【实测】/【经验】/【猜测】,猜测不当结论);采纳前先问「我们的前提一样吗」。
在授权修改范围内,将确定会重复的链路本轮参数化并跑通,省掉下轮重写;只读或禁止改脚本的任务只列候选。触发条件、落点与验收见 maintenance.md「五」第 1 条。
各项目的 .rankup/ 互不可见,登记表把可复用脚本索引到一处:node "<rankup-skill-dir>/scripts/registry.mjs" scan --roots <存放项目的目录> 整表重建,list 查看。它是 Skill 目录下的 registry.md,含项目名与绝对路径,因此被 .gitignore 排除,并由 scripts/validate-rankup.mjs 断言绝不能被 git 追踪。扫描根目录来自 --roots、RANKUP_PROJECT_ROOTS 或 ~/.rankup/config.json,绝不写死;只索引不复制。
查关键词、竞品、域名、页面、哥飞经验,按 seo-webcafe.md 检查并加载哥飞官方 Skill 包;未安装则按官方页面安装,再主动读取官方 gefei/SKILL.md 与对应专用 SKILL.md。专用 Skill 设了 disable-model-invocation: true,不会自动触发,必须主动加载。工具怎么调听官方,市场证据与闸门仍由 Rankup 判断。单一查询主 Agent 直接调;多个独立问题才分给子 Agent,收回后统一判读。
先装 opencli(npx skills add yan-labs/yan-skills --skill opencli -g -y):OpenCLI 本体装我们自己的构建,不是应用商店版(商店版默认前台抢标签页,失败不报错);opencli doctor 报扩展版本过低时照它说的做。一切浏览器动作用 opencli browser <描述性会话名> 的 dedicated 窗口,理由见 discipline.md 五。Web 项目的 Cloudflare 工具链:npm i -g cf、npx skills add yan-labs/yan-skills --skill cf-cli -g -y;现有项目继续用项目锁定的 Wrangler。
npx skills add yan-labs/yan-skills --skill rankup -g -y # 全局安装
npx skills update rankup -g -y # 全局更新
npx skills update rankup -p -y # 项目级更新发布版本记录在 skill.json;项目的启用时间、已安装版本和最近检查状态记录在 .rankup/skill-state.json。check-version.mjs 最多每 24 小时访问一次远端,只更新本 Skill;源码检出(仓库根有 .skill-source)或工作区有未提交修改时拒绝更新并报告原因,链接被换成实体目录时在仓库里跑 node scripts/link-skills.mjs 恢复。版本号规则见 maintenance.md「五」第 6 条。
.env,环境变量优先;细则见 discipline.md 十一。.rankup/、Git、测试或回复中保存真实密钥、token、密码、私钥、webhook secret、支付敏感数据或个人敏感信息。scripts/validate-rankup.mjs 断言。© yan-labs, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 264 other files (scripts, references) in rankup of yan-labs/yan-skills.
Open the folder on GitHubat commit 8e35596
Rankup next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Rankup this skillyan-labs/yan-skills | 213 | — | ~5.1k | Automated safety check: Notes | MIT | |
| Product Developmentgustavscirulis/snapgrid | 116 | 1 repos | ~1.1k | Automated safety check: Pass | Custom licence | |
| Submit Product Directories V2 Qualityflaqai/backlink_skills | 756 | — | ~2.2k | Automated safety check: Pass | MIT | |
| Global SEO Growthminhnv0807/ai-business-skills | 609 | — | ~5k | Automated safety check: Pass | MIT | |
| Competitor Analysispetera2c/simple-table | 229 | 2 repos | ~1.1k | Automated safety check: Pass | MIT | |
| SEO Backlink Auditrampstackco/claude-skills | 945 | — | ~2k | Automated safety check: Pass | MIT |
gustavscirulis/snapgrid
End-to-end product development for iOS/macOS apps. An agent skill from gustavscirulis/snapgrid.
flaqai/backlink_skills
SPD V2 Quality. An agent skill from flaqai/backlink_skills.
minhnv0807/ai-business-skills
Covers six SEO layers for a website: crawl and index audit, local SEO, search-intent content, AI search visibility, schema markup and backlink or directory distribution.
petera2c/simple-table
Analyze one competitor's organic footprint, ranking keywords, content themes, backlinks, and gaps.
rampstackco/claude-skills
Audit a backlink profile using Ahrefs MCP data: profile health, anchor text distribution, toxic link identification, lost link reclamation, and gap analysis against competitors.
minhnv0807/ai-business-skills
Khi nguoi dung muon tang traffic tu SEO, AI SEO, GEO/AEO, Google, ChatGPT, Perplexity, schema, structured data, llms.txt, programmatic SEO, local SEO, directory submissions, Product Hunt…
yan-labs/yan-skills
A skill your agent uses for backlink work - finding, qualifying, submitting, analyzing or verifying backlinks, directory and blog-comment placements, competitor link sources, anchors, toxic links…
yan-labs/yan-skills
当用户要用本机已登录 Chrome 操作或验收网页(含测试自己的站、E2E、截图)、读取登录后台、抓取无 API 的数据、填表,或排查 OpenCLI doctor、会话、adapter 故障,或明确提到 opencli、浏览器自动化、标签页被抢时使用。公开文本先用 HTTP;话题调研交 agent-reach;SEO 与选词决策交 rankup;外链选点、提交与…
yan-labs/yan-skills
A skill your agent uses when a skill or slash command is missing, duplicated, not loading, or inconsistent between .agents/skills and .claude/skills; includes “skill 没生效”“检查 skill 链接” and broken…
yan-labs/yan-skills
使用本机 fleet 分派 Codex GPT-6.1 Sol、Gemini、Grok 或 JEV 任务,或调用 Kollab 图片/视频/音频/多模态能力时使用;包括用户点名 agent-fleet、Nano Banana、nanobanana、香蕉、便宜模型、多模型并行,用户说“让 Codex 或 GPT-6.1 Sol 做某事”的编码、调研与 review 派单,以及按全局…
yan-labs/yan-skills
当用户明确要求直接操作 Codex CLI 作为后台代理(自选 sandbox、codex review、apply、resume 等原生命令,或用 git worktree 并行布置多个 worker)做代码分析、编辑、审查时使用。普通的“让 Codex 或 GPT-6 写代码、调研、review”派单走 agent-fleet(fleet code);图片生成与网站视觉素材走…
yan-labs/yan-skills
A skill your agent uses when the user asks for Cloudflare's cf CLI, cf cli search, or cloudflare.config.ts to inspect resources, manage zones or DNS, attach domains, or deploy.
独立开发者的项目全生命周期管理:需求验证、选词选品、建站或做原生 App、上线接入、SEO/GEO 获客、支付变现、监控迭代与跨会话接力。以下情况使用:用户提到 rankup 或 /rankup(check、review、init;doctor 仅指整理项目 .rankup 记录);当前目录或工作区有…. Rankup is an agent skill from yan-labs/yan-skills.
Rankup fits situations like: tasks that involve Link building; tasks that involve Image generation; tasks that involve PRD writing.
Run `npx skills add yan-labs/yan-skills --skill rankup -a claude-code`. Or copy the skill folder (rankup in yan-labs/yan-skills) into .claude/skills/rankup in your project. Claude Code loads it when a task matches its description.
Run `npx skills add yan-labs/yan-skills --skill rankup -a codex`. Or copy the skill folder (rankup in yan-labs/yan-skills) into .agents/skills/rankup in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add yan-labs/yan-skills --skill rankup -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/rankup, .gemini/skills/rankup, .github/skills/rankup and .opencode/skills/rankup in your project.
Going by SKILL.md and its folder, Rankup needs the command-line tools its instructions call (npx, node, git and npm).
SKILL.md names 1 domain. As links in the text: seo.web.cafe. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
Rankup is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 5.1k tokens (SKILL.md is roughly 20k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 413k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Rankup: Product Development (gustavscirulis/snapgrid, 116 stars), Submit Product Directories V2 Quality (flaqai/backlink_skills, 756 stars), Global SEO Growth (minhnv0807/ai-business-skills, 609 stars) and Competitor Analysis (petera2c/simple-table, 229 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
yan-labs (a GitHub user) maintains it in yan-labs/yan-skills, which has 213 GitHub stars. The repository holds 9 skills in this directory. The repository was last updated on October 10, 2026.
Source: yan-labs/yan-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.