Fix internal linking, orphan pages and topic hubs, with your agent
Audits and fixes your site’s internal link graph, orphan pages, topic hubs, crawl depth and BreadcrumbList schema in code, then verifies the updated structure on the live site with clickd.
Add it, from the top of the site’s repositorynpx skills add subinternal/clickd --skill clickd-site-architecture
Then ask your agentFind all orphan and under-linked pages on example.com with clickd, connect them into our hub pages with descriptive anchor text, and add breadcrumbs.
For developers and SEO owners on growing sites with dozens or hundreds of pages
Four steps from link graph audit to three-click depth
Your agent maps every page’s incoming links and crawl depth, connects orphan routes into topic hubs, and wires visible breadcrumbs.
1Map the internal link graph and find orphan or deep pagesclickd_links
2Group pages into clear hubs and spokesclickd_topics
3Wire visible breadcrumbs and BreadcrumbList schema
4Rescan and verify the link graph with clickdclickd_rescan
Internal link graph before and after
Fixing orphan routes and deep crawl paths in templates without breaking working URLs.
example.com link & depth read
112 indexable pages in sitemap; checked with clickd_links and clickd_structure.
Sample
Orphan pages (0 links in)14 sitemap pages had 0 incoming internal <a href> links0 orphan pages; every route linked from its section hub and 2+ sibling pages
Crawl depth from /Max depth 6 clicks; 29 pages sat 4+ clicks from homeMax depth 3 clicks across all 112 indexable pages
Internal link hygiene18 internal links pointed to 301 redirects or generic “read more”Direct 200 canonical hrefs with descriptive topic anchor text
BreadcrumbsNo breadcrumb trail on /guides/* or /services/* templatesVisible <nav aria-label="Breadcrumb"> + matching BreadcrumbList JSON-LD
How clickd measures site architecture
Read from the real links crawlers follow in server-rendered HTML.
clickd_links & clickd_structureLists incoming and outgoing internal links per page, orphan pages, redirect hops and click depth from home.
clickd_topics & clickd_mapShows how pages cluster by subject so hub-and-spoke links follow the site’s real topics.
clickd_rescan & clickd_changesConfirms orphan-page, internal-link and breadcrumb findings clear on the live site.
When to reach for it
Reach for it when
Important pages sit in the sitemap with zero or one incoming internal link.1
Deep guides, product pages or city pages take four or more clicks to reach from the home page.
You want clean hub-and-spoke topic clusters and BreadcrumbList markup across deep templates.3
Leave it when
The site only has a handful of pages and every one is already linked in the main header and footer.
You want to rename working URLs across the site without a technical reason.
More things to ask your agent
Everyday prompts that load the Site architecture skill in Claude Code, Codex or Cursor.
Eliminate orphan pagesRun clickd_links on example.com, list every sitemap page with 0 or 1 incoming internal links, and link them from our hubs.
Flatten deep crawl pathsFind every page more than 3 clicks from the home page with clickd_structure and add hub-and-spoke navigation.
Add BreadcrumbList across templatesAdd a visible breadcrumb nav and matching BreadcrumbList JSON-LD to our /guides/* and /services/* templates.
Site architecture skill vs the alternatives
Why pairing an open repository skill with a live crawler beats manual audits and prompt-only checklists.
How it reads your siteBy hand: Manual spot-checks in a browser · Prompt-only skill: Reads local source files only; misses live headers, redirects and JS shellsReads the live URL, HTTP headers, raw HTML and Search Console via clickd MCP
How changes are madeBy hand: Copy-pasting between docs, spreadsheets and editors · Prompt-only skill: Suggests generic copy in chat with no repo contextEdits your actual templates, routes and schema in the repository
How fixes are provedBy hand: Checking pages one by one after deploy · Prompt-only skill: Agent grades its own homework with no live checkRuns clickd_rescan and clickd_changes on the deployed site to prove rules passed
Skills to use alongside it
Included when you run npx skills add subinternal/clickd.
Technical SEOAudits and fixes crawlability, indexing, robots.txt, XML sitemaps, canonicals, redirect chains, server-rendered HTML and security headers in your site’s codebase, then checks every fix live with clickd.See skill
Page systemBuilds a site’s page templates and shared components so crawlers and assistants read every page well, and proves each pattern with clickd. About structure and spacing, not color, type or brand.See skill
Content briefTurns competitor topic gaps and customer questions from clickd’s content brain into fact-backed briefs and finished pages in your repo, complete with question headings, internal links and JSON-LD schema.See skill
Questions about the Site architecture skill
Will Site architecture rename our URLs and break bookmarks?
No. It prioritizes fixing internal links, hub indexes and breadcrumbs first, and only recommends changing a URL when it is broken or duplicated (always with a 301 redirect).
How many incoming internal links should an indexable page have?
Every indexable page in your sitemap should have at least 2 crawlable <a href> links from relevant hub or sibling pages and sit within 3 clicks of the home page.
Install it by hand
The line at the top adds it for any agent. Or fetch its 1 files where your agent looks. Your agent also needs clickd connected over MCP: connecting an agent shows how.
Loads skills from .claude/skills/ in the repository, for everyone who works in it, or ~/.claude/skills/ for every project on your machine. Claude uses it when a request matches its description, or type /clickd-site-architecture.
Puts the files in .claude/skills/clickd-site-architecture/.
Loads skills from .agents/skills/ in the repository and its parent folders, or ~/.agents/skills/ for every project. Codex uses it when a task matches its description, or mention it as $clickd-site-architecture; /skills lists them.
Puts the files in .agents/skills/clickd-site-architecture/.
Loads skills from .cursor/skills/ or .agents/skills/ in the repository, or ~/.cursor/skills/ for every project. It also reads .claude/skills/, so one copy there serves both. The agent uses it when it is relevant, or type / in Agent chat and pick it.
Puts the files in .cursor/skills/clickd-site-architecture/.
An agent that does not load skills can still follow the file. Save the folder in the repository and point its standing instructions at it, with a line in AGENTS.md, which many agents read. The agent reads AGENTS.md when it starts, and follows the file when the task matches.
SKILL.md stays short and sends your agent to a reference when a step needs it.
SKILL.mdThe four steps from mapping the internal link graph and finding orphan or deep pages to wiring hub-and-spoke links, adding BreadcrumbList schema, and verifying live with clickd.87 lines
Each is in a ## Sources list in the file that cites it, by the page’s own title.
Google Search Central
1Link Best Practices for Google, Google Search Central.Using crawlable `<a href>` elements and descriptive anchor text so search engines understand what linked pages are about.
2Organize Your Site Hierarchy, Google Search Central.Structuring directories, navigation and internal links for people and crawlers.
3Breadcrumb Structured Data, Google Search Central.Marking up `BreadcrumbList` so search results can show a page's position in the site hierarchy.
4Keep a Simple URL Structure, Google Search Central.Designing readable, lowercase URLs with words rather than opaque IDs.
Agent skills format and clients
5Extend Claude with skills, Anthropic, Claude Code docs.Where Claude Code loads skills from, and how one is called.
6Build skills, OpenAI, ChatGPT Learn.Where Codex loads skills from, and how one is called.
7Agent Skills, Cursor docs.Where Cursor loads skills from, and how one is called.
8Specification, Agent Skills.The format every skill here keeps: the folder, the frontmatter and its limits.