← SkillSafe / pSEO Desk

Five hundred pages, or one page five hundred times?

Paste the page template behind your programmatic SEO pages - with {{column}} placeholders - and the data rows that fill it: integration pages, tool-vs-tool comparisons, city pages, glossary terms, directories. Your browser renders every page and measures what the programmatic-seo skill warns about: how much of each page is the template's own text, which pages are near-duplicates of each other, duplicate URLs and titles, the same comparison in both orders, placeholders that print blank, thin pages, no internal links, no schema, rows nobody searches for. Free, before you sign in, and nothing is uploaded. Then the audit says whether to launch, and the template lane writes a stronger template that the page renders over your data again.

The businesses, tools and numbers in the examples are invented. Two are audits (the integration one also carries its template, one click after the audit) and one is a template; each comes with a saved model reply, so you can see the whole page without signing in or spending a credit.

Check Build

nothing pasted yet
Drop an HTML, Markdown, text, Handlebars, Mustache or Liquid file (read in your browser), or
nothing pasted yet
Drop a CSV, TSV or Markdown table (read in your browser), or
optional: business, goal, data source, demand
Paste a template, data rows or both to price the run.

What this does, and what it does not

Programmatic SEO fails quietly. A template that only swaps a city or a tool name produces hundreds of pages that say the same thing, and search engines treat them as thin or doorway pages. A placeholder with no data prints a gap on every page; a duplicate row makes two pages fight for one URL; "A vs B" and "B vs A" compete for one keyword; pages with no links to a hub end up orphaned; rows nobody searches for waste crawl budget. The skill's answer is unique value per page from data you own, conditional content driven by that data, clean subfolder URLs, hub-and-spoke linking, schema, and an indexation plan that holds back the thin rows. This page renders your pages, measures each one, and raises the mistakes the skill warns about. Its word-count and template-text lines are its own scale, shown next to every number, not rules from a search engine.

The paid lanes see your template (the middle is cut past 8,000 characters, and the page says so), the header and as many whole data rows as fit in 5,000 characters, your context (past 4,000) and the page's scan with three sample pages. Your input stays in this browser unless you run a lane. Every reply is held to your input: quotes are checked word for word, and no count, price, volume or ranking may be invented - template copy uses your columns or [placeholders] for claims only you can make. It does not research keywords or fetch search volumes; bring those as a column. Derived from the agent skill @coreyhaines31/programmatic-seo by Corey Haines (coreyhaines31/marketingskills, MIT): its core principles, the 12 playbooks and how to choose one, the implementation framework (keyword pattern, data requirements, template design, internal linking, indexation), the pre-launch checklist and the common mistakes. Sources are listed in llms.txt.