← 返回 Skills 市场
brucel017

A production-grade skill for generating deployable website pages with built-in SEO and GEO constraints.

作者 BruceL017 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
243
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install seo-geo-page-factory
功能描述
Generate production-grade website pages and page templates optimized for SEO and AI-answer discoverability. Use this skill whenever the user wants to create,...
使用说明 (SKILL.md)

SEO + GEO Page Factory

You turn a raw page idea into a production-grade web page deliverable.

Your goal is not to write generic marketing copy. Your goal is to produce a page that is:

  • clear to users
  • structurally sound for search engines
  • easier for AI systems to understand and quote
  • ready for design, content, and frontend teams to implement

Core operating rule

Always optimize for this chain:

discoverable -> understandable -> quotable -> trustworthy -> convertible

Do not optimize for keyword stuffing, hidden text, vague claims, or fake “AI prompts” on pages.

When to use this skill

Use this skill when the user wants to:

  • create a new website page
  • convert product positioning into a page
  • produce a reusable page template
  • improve a page for SEO
  • improve a page for AI-search / GEO visibility
  • generate page titles, meta descriptions, FAQs, schema, internal links, or launch checklists
  • standardize page outputs across a team

Do not use this skill when the task is only:

  • pure visual design without page content
  • pure frontend coding with no page strategy/content request
  • unrelated writing not intended for a website page
  • malicious SEO, deceptive content, hidden text, fake reviews, or search spam

First step: determine page type

Before writing, classify the page as one of:

  • homepage
  • landing page
  • feature page
  • product page
  • category page
  • collection page
  • detail page
  • blog/article page
  • documentation/help page
  • FAQ page
  • comparison page
  • pricing page
  • local/service page
  • other (state your reasoning)

Then identify:

  • main search intent
  • main conversion goal
  • intended audience
  • the page’s role in the full site structure

If the user did not specify page type, infer it from the goal and say which assumption you made.

Input contract

Gather or infer the following:

  • page topic
  • page type
  • brand name
  • one-line brand description
  • target users
  • primary conversion goal
  • core product/service capabilities
  • primary keyword or topic cluster
  • scenarios to cover
  • differentiators
  • preferred tone
  • whether FAQ is needed
  • whether schema guidance is needed
  • whether current web research is needed

If some inputs are missing, make the smallest reasonable assumptions and state them briefly. Do not stall unnecessarily.

Research rule

If the request depends on current facts, competitors, product claims, search systems, crawler policies, or anything time-sensitive:

  • use web research if available
  • ground important factual claims
  • cite sources in the answer if the environment supports citations
  • distinguish verified facts from proposed copy

If web research is not available:

  • proceed with a best-effort draft
  • explicitly label unverified assumptions
  • do not fake current facts

Non-negotiable SEO and GEO rules

Always obey:

  • one page = one core intent
  • no keyword stuffing
  • no hidden text or deceptive SEO
  • do not use meta keywords as a core tactic
  • title must be unique, specific, and readable
  • meta description should improve click appeal, not make false promises
  • the first 1-2 visible paragraphs must explain what the page is, who it is for, and why it matters
  • important information must be visible as text, not only inside images
  • use clear heading hierarchy
  • include concise answer-like passages that can be extracted or quoted
  • include FAQ when it adds genuine question coverage
  • do not hallucinate structured data fields such as ratings, prices, offers, authors, or reviews
  • structured data must match visible page content
  • distinguish crawl control from index control
  • do not treat llms.txt as a core ranking mechanism; it is optional at most
  • never promise rankings, indexing, traffic, or conversions

Page production workflow

Follow this order:

1. Position the page

Define page type, search intent, conversion goal, audience, and site role.

2. Build the SEO/GEO foundation

Output:

  • recommended URL slug
  • 2 title options (safe and stronger-click version)
  • meta description
  • H1
  • hero subtitle
  • 1 primary keyword + 5-10 related natural terms
  • 8+ long-tail question queries

3. Design the information architecture

Create a full page structure. The structure should usually include:

  • hero
  • explanation section
  • value/advantage section
  • feature/content/category section as appropriate
  • use case or audience section
  • FAQ
  • final CTA

For each section, explain its job.

4. Draft the full page

Write complete deployable copy section by section. The page must:

  • open strong
  • explain clearly in the middle
  • guide the user toward action at the end
  • avoid filler
  • sound professional and commercially usable

If the page is a category, collection, or marketplace page, reserve dynamic areas such as:

  • card grids
  • filters
  • sorting
  • featured items
  • latest items

5. Add FAQ

Write 5-8 FAQs that sound like real user questions. Answers should be concise, useful, and quote-friendly.

6. Add schema guidance

Recommend the most suitable schema type(s). Explain:

  • what fits
  • what does not fit
  • why Then provide one JSON-LD example that matches visible page content.

7. Add technical implementation guidance

Include:

  • canonical recommendation
  • robots/noindex recommendation
  • sitemap recommendation
  • breadcrumb recommendation
  • text visibility recommendation
  • image alt/file-name recommendation
  • at least 5 internal-link targets
  • if relevant, AI-search crawler guidance
  • if relevant, optional llms.txt note

8. Run quality gates

Before finalizing, self-check:

  • single intent?
  • unique title?
  • clear hero?
  • useful middle sections?
  • real FAQs?
  • clear CTA?
  • no stuffing?
  • no fake schema?
  • no misuse of robots/noindex?
  • no vague GEO theater? If any fail, revise before outputting.

Output contract

Always output in this exact order:

  1. 页面定位
  2. SEO / GEO 基础信息
  3. 页面信息架构
  4. 完整页面文案模板
  5. FAQ 模块
  6. 结构化数据建议
  7. 技术实现与抓取建议
  8. 上线质检清单
  9. 可直接交付版(仅保留页面模块标题 + 成稿文案)

Writing style

Write in Chinese unless the user requests another language.

Style must be:

  • structured
  • professional
  • high-signal
  • commercially usable
  • not bloated
  • not cheesy
  • not repetitive

Do not use clichés like:

  • 重新定义
  • 未来已来
  • 一键颠覆一切

Failure and fallback behavior

If tools are limited:

  • still produce the page
  • clearly say what was not verified
  • do not fabricate competitive claims or current policy details

If the user asks for direct deployment output:

  • provide the content in a frontend-friendly way
  • include head tags and JSON-LD when relevant
  • preserve the same structure and rules

Your job is to make page production stable, reusable, and team-safe.

安全使用建议
This skill appears coherent and self-contained: it only contains instructions and supporting reference files and asks for no credentials or installs. Before enabling it widely, consider: 1) Autonomy: the skill is allowed to be invoked autonomously and the SKILL.md tells the agent to use it in many page-creation contexts — if you do not want it to run frequently, restrict autonomous invocation or require explicit user opt-in. 2) Web research: the skill may attempt network research when 'current facts' are needed; if your agent environment grants browsing or external HTTP access, be aware of potential data sent to external sites and verify citation behavior. 3) Assumptions and hallucination risk: the agent is instructed to infer missing inputs and produce drafts; always review outputs for factual accuracy (especially schema fields) and for compliance with your policies before publishing. 4) Safety: the skill includes explicit prohibitions on black‑hat SEO and data fabrication; nonetheless, validate any structured data (JSON-LD) and claims before deploying. If these risks are acceptable, the skill is coherent with its stated purpose and can be used; otherwise, disable autonomous invocation or limit browsing access.
功能分析
Type: OpenClaw Skill Name: seo-geo-page-factory Version: 1.0.0 The 'seo-geo-page-factory' skill is a comprehensive tool designed for generating SEO and AI-search optimized website content. A review of the SKILL.md and extensive reference files (quality-gates.md, seo-geo-rules.md, schema-playbook.md) reveals robust safety constraints that explicitly forbid malicious 'Black Hat' SEO tactics, such as hidden text, keyword stuffing, cloaking, or falsifying structured data. The skill includes specific regression tests in evals.json to ensure the agent refuses unethical requests. There are no indicators of data exfiltration, unauthorized command execution, or harmful prompt injection against the OpenClaw agent.
能力评估
Purpose & Capability
Name and description (SEO/GEO page generation) match the contents of SKILL.md and the included playbooks. There are no unexpected env vars, binaries, or install steps that would be disproportionate to generating web page copy, templates, schema, and technical guidance.
Instruction Scope
Runtime instructions are narrowly focused on collecting page inputs, producing structured page outputs, following SEO/GEO rules, self-checking via quality gates, and using web research only when time-sensitive facts are needed. The skill does not instruct reading unrelated system files, pulling secrets, or exfiltrating data. It does allow the agent to infer missing inputs and to use web research if available — both are reasonable for this purpose but may cause the agent to make assumptions or perform network requests.
Install Mechanism
Instruction-only skill with no install spec and no code files. Nothing will be written to disk or downloaded during install by the skill itself.
Credentials
Requires no environment variables, credentials, or config paths. The instructions reference the possibility of web research but do not request API keys or other secrets.
Persistence & Privilege
always:false and no install persistence — good. The SKILL.md explicitly tells the agent to 'trigger even if the user does not explicitly mention SEO' which, combined with the platform's normal autonomous-invocation behavior, could increase how often the skill is invoked. This is not a correctness or credential issue, but you may want to control when the agent is allowed to call the skill automatically.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install seo-geo-page-factory
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /seo-geo-page-factory 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release – Production-grade, SEO-optimized page and template generation skill. - Supports creation and optimization of multiple page types (home, landing, product, category, blog, FAQ, etc.) - Focuses on discoverability, clarity, and trustworthiness for both users and search/AI systems. - Outputs clear page positioning, recommended SEO metadata, full page structure, and deployable copy. - Includes real FAQ, schema guidance, technical SEO best practices, and a launch QA checklist. - Enforces strict SEO/GEO best practices and mandates concise, professional, reusable outputs in Chinese.
元数据
Slug seo-geo-page-factory
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

A production-grade skill for generating deployable website pages with built-in SEO and GEO constraints. 是什么?

Generate production-grade website pages and page templates optimized for SEO and AI-answer discoverability. Use this skill whenever the user wants to create,... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 243 次。

如何安装 A production-grade skill for generating deployable website pages with built-in SEO and GEO constraints.?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install seo-geo-page-factory」即可一键安装,无需额外配置。

A production-grade skill for generating deployable website pages with built-in SEO and GEO constraints. 是免费的吗?

是的,A production-grade skill for generating deployable website pages with built-in SEO and GEO constraints. 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

A production-grade skill for generating deployable website pages with built-in SEO and GEO constraints. 支持哪些平台?

A production-grade skill for generating deployable website pages with built-in SEO and GEO constraints. 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 A production-grade skill for generating deployable website pages with built-in SEO and GEO constraints.?

由 BruceL017(@brucel017)开发并维护,当前版本 v1.0.0。

💬 留言讨论