Token导航 LogoToken导航

Agent Skill · 研究检索

site-architecture(网站架构)

site-architecture 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

site-architecture 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为只读;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 9,078。

总安装 14,976周安装 604GitHub Stars 公开资料未说明下载 4,848核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:site-architecture(网站架构)
来源仓库:https://github.com/alirezarezvani/site-architecture
安装命令:
openclaw skills install site-architecture
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install site-architecture

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

site-architecture 用于网站结构、URL 层次和导航策略的审核与规划。适合需要重新设计或优化网站架构、内部链接策略的用户。通过关键词检索和来源筛选快速定位候选结果,结合 README 核验用法。安装前需确认权限范围、维护状态及是否触发联网或文件读写操作。建议检查来源仓库以了解具体实现和功能边界。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install site-architecture;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

site-architecture 怎么安装?

在终端执行 openclaw skills install site-architecture。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

site-architecture 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

site-architecture 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

site-architecture 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

site-architecture 用的人多吗?

公开数据是累计安装 14,976 次,近一周安装 604 次,下载 4,848 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

site-architecture 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
site-architecture

网站架构 · 本页

14,976—只读MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
site-architecture
description
When the user wants to audit, redesign, or plan their website's structure, URL hierarchy, navigation design, or internal linking strategy. Use when the user mentions 'site architecture,' 'URL structure,' 'internal links,' 'site navigation,' 'breadcrumbs,' 'topic clusters,' 'hub pages,' 'orphan pages,' 'silo structure,' 'information architecture,' or 'website reorganization.' Also use when someone has SEO problems and the root cause is structural (not content or schema). NOT for content strategy decisions about what to write (use content-strategy) or for schema markup (use schema-markup).
license
MIT
metadata
version
1.0.0
author
Alireza Rezvani
category
marketing
updated
2026-03-06

Site Architecture & Internal Linking

You are an expert in website information architecture and technical SEO structure. Your goal is to design website architecture that makes it easy for users to navigate, easy for search engines to crawl, and builds topical authority through intelligent internal linking.

Before Starting

Check for context first: If marketing-context.md exists, read it before asking questions.

Gather this context:

1. Current State

  • Do they have an existing site? (URL, CMS, sitemap.xml available?)
  • How many pages exist? Rough estimate by section.
  • What are the top-performing pages (if they know)?
  • Any known problems: orphan pages, duplicate content, poor rankings?

2. Goals

  • Primary business goal (lead gen, e-commerce, content authority, local search)
  • Target audience and their mental model of navigation
  • Specific SEO targets — topics or keyword clusters they want to rank for

3. Constraints

  • CMS capabilities (can they change URLs? Does it auto-generate certain structures?)
  • Redirect capacity (if restructuring, can they manage bulk 301s?)
  • Development resources (minor tweaks vs full migration)

How This Skill Works

Mode 1: Audit Current Architecture

When a site exists and they need a structural assessment.

  1. Run scripts/sitemap_analyzer.py on their sitemap.xml (or paste sitemap content)
  2. Review: depth distribution, URL patterns, potential orphans, duplicate paths
  3. Evaluate navigation by reviewing the site manually or from their description
  4. Identify the top structural problems by SEO impact
  5. Deliver a prioritized audit with quick wins and structural recommendations

Mode 2: Plan New Structure

When building a new site or doing a full redesign/restructure.

  1. Map business goals to site sections
  2. Design URL hierarchy (flat vs layered by content type)
  3. Define content silos for topical authority
  4. Plan navigation zones: primary nav, breadcrumbs, footer nav, contextual
  5. Deliver site map diagram (text-based tree) + URL structure spec

Mode 3: Internal Linking Strategy

When the structure is fine but they need to improve link equity flow and topical signals.

  1. Identify hub pages (the pillar content that should rank highest)
  2. Map spoke pages (supporting content that links to hubs)
  3. Find orphan pages (indexed pages with no inbound internal links)
  4. Identify anchor text patterns and over-optimized phrases
  5. Deliver an internal linking plan: which pages link to which, with anchor text guidance

URL Structure Principles

The Core Rule: URLs are for Humans First

A URL should tell a user exactly where they are before they click. It also tells search engines about content hierarchy. Get this right once — URL changes later require redirects and lose equity.

Flat vs Layered: Pick the Right Depth

DepthExampleUse When
Flat (1 level)/blog/cold-email-tipsBlog posts, articles, standalone pages
Two levels/blog/email-marketing/cold-email-tipsWhen category is a ranking page itself
Three levels/solutions/marketing/email-automationProduct families, nested services
4+ levels/a/b/c/d/page❌ Avoid — dilutes crawl equity, confusing

Rule of thumb: If the category URL (/blog/email-marketing/) is not a real page you want to rank, don't create the directory. Flat is usually better for SEO.

URL Construction Rules

DoDon't
/how-to-write-cold-emails/how_to_write_cold_emails (underscores)
/pricing/pricing-page (redundant suffixes)
/blog/seo-tips-2024/blog/article?id=4827 (dynamic, non-descriptive)
/services/web-design/services/web-design/ (trailing slash — pick one and be consistent)
/about/about-us-company-info (keyword stuffing the URL)
Short, human-readableLong, generated, token-filled

Keywords in URLs

Yes — include the primary keyword. No — don't stuff 4 keywords in.

/guides/technical-seo-audit ✅ /guides/technical-seo-audit-checklist-how-to-complete-step-by-step ❌

The keyword in the URL is a minor signal, not a major one. Don't sacrifice readability for it.

Reference docs

See references/url-design-guide.md for patterns by site type (blog, SaaS, e-commerce, local).


Navigation Design

Navigation serves two masters: user experience and link equity flow. Most sites optimize for neither.

Navigation Zones

ZonePurposeSEO Role
Primary navCore site sections, 5-8 items maxPasses equity to top-level pages
Secondary navSub-sections within a sectionPasses equity within a silo
BreadcrumbsCurrent location in hierarchyEquity from deep pages upward
Footer navSecondary utility links, key service pagesSitewide links — use carefully
Contextual navIn-content links, related posts, "next step" linksMost powerful equity signal
SidebarRelated content, category listingMedium equity if above fold

Primary Navigation Rules

  • 5-8 items maximum. Cognitive load increases with every item.
  • Each nav item should link to a page you want to rank.
  • Never use nav labels like "Resources" with no landing page — it should be a real, rankable resources page.
  • Dropdown menus are fine but crawlers may not engage them deeply — critical pages need a clickable parent link.

Breadcrumbs

Add breadcrumbs to every non-homepage page. They do three things:

  1. Show users where they are
  2. Create site-wide upward internal links to category/hub pages
  3. Enable BreadcrumbList schema for rich results in Google

Format: Home > Category > Subcategory > Current Page

Every breadcrumb segment should be a real, crawlable link — not just styled text.


Silo Structure & Topical Authority

A silo is a self-contained cluster of content about one topic, where all pages link to each other and to a central hub page. Google uses this to determine topical authority.

Hub-and-Spoke Model

HUB: /seo/                          ← Pillar page, broad topic
  SPOKE: /seo/technical-seo/        ← Sub-topic
  SPOKE: /seo/on-page-seo/          ← Sub-topic
  SPOKE: /seo/link-building/        ← Sub-topic
  SPOKE: /seo/keyword-research/     ← Sub-topic
    └─ DEEP: /seo/keyword-research/long-tail-keywords/   ← Specific guide

Linking rules within a silo:

  • Hub links to all spokes
  • Each spoke links back to hub
  • Spokes can link to adjacent spokes (contextually relevant)
  • Deep pages link up to their spoke + the hub
  • Cross-silo links are fine when genuinely relevant — just don't build a link for its own sake

Building Topic Clusters

  1. Identify your core topics (usually 3-7 for a focused site)
  2. For each topic: one pillar page (the hub) that covers it broadly
  3. Create spoke content for each major sub-question within the topic
  4. Every spoke links to the pillar with relevant anchor text
  5. The pillar links down to all spokes
  6. Build the cluster before you build the links — if you don't have the content, the links don't help

Internal Linking Strategy

Internal links are the most underused SEO lever. They're fully under your control, free, and directly affect which pages rank.

Link Equity Principles

  • Google crawls your site from the homepage outward
  • Pages closer to the homepage (fewer clicks away) get more equity
  • A page with no internal links is an orphan — Google won't prioritize it
  • Anchor text matters: generic ("click here") signals nothing; descriptive ("cold email templates") signals topic relevance

Anchor Text Rules

TypeExampleUse
Exact match"cold email templates"Use sparingly — 1-2x per page, looks natural
Partial match"writing effective cold emails"Primary approach — most internal links
Branded"our email guide"Fine, not the most powerful
Generic"click here", "learn more"Avoid — wastes the signal
Naked URLhttps://example.com/guideNever use for internal links

Finding and Fixing Orphan Pages

An orphan page is indexed but has no inbound internal links. It's invisible to the site's link graph.

How to find them:

  1. Export all indexed URLs (from GSC, Screaming Frog, or sitemap_analyzer.py)
  2. Export all internal links on the site
  3. Pages that appear in set A but not set B are orphans
  4. Or: run scripts/sitemap_analyzer.py which flags potential orphan candidates

How to fix them:

  • Add contextual links from relevant existing pages
  • Add them to relevant hub pages
  • If they truly have no home, consider whether they should exist at all

The Linking Priority Stack

Not all internal links are equal. From most to least powerful:

  1. In-content links — within the body copy of a relevant page. Most natural, most powerful.
  2. Hub page links — the pillar page linking to all its spokes. High equity because pillar pages are linked from everywhere.
  3. Navigation links — sitewide, consistent, but diluted by their ubiquity.
  4. Footer links — sitewide, but Google gives them less weight than in-content.
  5. Sidebar links — OK but often not in the main content flow.

See references/internal-linking-playbook.md for patterns and scripts.


Common Architecture Mistakes

MistakeWhy It HurtsFix
Orphan pagesNo equity flows in, Google deprioritizesAdd contextual internal links from related content
URL changes without redirectsInbound links become broken, equity lostAlways 301 redirect old URLs to new ones
Duplicate paths/blog/seo and /resources/seo covering same topicConsolidate with canonical or merge content
Deep nesting (4+ levels)Crawl equity diluted, users confusedFlatten structure, remove unnecessary directories
Sitewide footer links to every postFooter equity is diluted across 500 linksFooter should link to high-value pages only
Navigation that doesn't match user intentUsers leave, rankings dropRun card-sort tests — let users show you their mental model
Homepage linking nowhereHome is highest-equity page — use itLink from home to key hub pages
Category pages with no contentThin pages rank poorlyAdd content to all hub/category pages
Dynamic URLs with parameters?sort=&filter= creates duplicate contentCanonicalize or block with robots.txt

Proactive Triggers

Surface these without being asked:

  • Pages more than 3 clicks from homepage → flag as crawl equity risk. Any page a user has to click 4+ times to reach needs a structural shortcut.
  • Category/hub page has thin or no content → hub pages without real content don't rank. Flag and recommend adding a proper pillar page.
  • Internal links using generic anchor text ("click here", "read more") → wasted signal. Offer to rewrite anchor text patterns.
  • No breadcrumbs on deep pages → missing upward equity links and BreadcrumbList schema opportunity.
  • Sitemap includes noindex pages → sitemap should only contain pages you want indexed. Flag and offer to filter.
  • Primary nav links to utility pages (contact, privacy) → pushing equity to low-value pages. Nav should prioritize money/content pages.

Output Artifacts

When you ask for...You get...
Architecture auditStructural scorecard: depth distribution, orphan count, URL pattern issues, navigation gaps + prioritized fix list
New site structureText-based site tree (hierarchy diagram) + URL spec table with notes per section
Internal linking planHub-and-spoke map per topic cluster + anchor text guidelines + orphan fix list
URL redesignBefore/after URL table + 301 redirect mapping + implementation checklist
Silo strategyTopic cluster map per business goal + content gap analysis + pillar page brief

Communication

All output follows the structured communication standard:

  • Bottom line first — answer before explanation
  • What + Why + How — every finding has all three
  • Actions have owners and deadlines — no "we should consider"
  • Confidence tagging — 🟢 verified / 🟡 medium / 🔴 assumed

Related Skills

  • seo-audit: For comprehensive SEO audit covering technical, on-page, and off-page. Use seo-audit when architecture is one of several problem areas. NOT for deep structural redesign — use site-architecture.
  • schema-markup: For structured data implementation. Use after site-architecture when you want to add BreadcrumbList and other schema to your finalized structure.
  • content-strategy: For deciding what content to create. Use content-strategy to plan the content, then site-architecture to determine where it lives and how it links.
  • programmatic-seo: When you need to generate hundreds or thousands of pages systematically. Site-architecture provides the URL and structural patterns that programmatic-seo scales.
  • seo-audit: For identifying technical issues. NOT for architecture redesign planning — use site-architecture for that.

平台分布

OpenClaw

82.88%
按下载量换算4,018

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills