seo-sxo checks whether a page deserves to rank for a keyword given what Google shows. It renders the target page, searches the keyword, classifies the top ten results into eight page types, and finds the dominant type and a mismatch severity.
It then derives user stories from People Also Ask questions, ad copy and related searches, scores the page across seven dimensions for a 100-point SXO Gap Score, rates four to seven personas on relevance, clarity, trust and action, and can output current and target wireframes.
What You Get
- A SERP landscape with the dominant page type and consensus level
- A page-type verdict from aligned to critical mismatch
- Three to five user stories that cite the SERP signal behind each
- Persona cards with scores and improvement notes
- IST and SOLL wireframes with concrete placeholders on request
Common Use Cases
- Explain why a well-optimized blog post cannot beat product pages
- Decide whether to build a comparison page or a tool for a keyword
- Find which audience segment a landing page serves worst
- Brief a designer with a target page outline
What Makes It Unique
It starts from the SERP and works backward to the page. Four reference files define the page type taxonomy, user story format, persona scoring and wireframe templates, and the wireframe step demands specific placeholders such as a pricing CTA with a link target.
Under The Hood
The skill runs the plugin’s render_page.py and parse_html.py through a launcher, which fetch the target page over the network with an SSRF guard and may render it in a headless browser. It then uses web search to read the top ten results for the keyword, and can call DataForSEO after a cost estimate and confirmation. It reads four reference files on demand and writes no files unless the user asks for one. The frontmatter declares none of this access. Fetched pages and search results are untrusted data, and the skill tells the agent not to change browser sandbox settings itself.