/suede-visibility-grader · Website visibility and CTA grading
No live inspection caps the page at C. A broken CTA caps it at D.
Each lane is graded A to F from inspection evidence, and a false published statement caps the page at D or drops it to F. Findability, first-screen clarity, CTA pull, proof and trust, AI readability, and design signal, with the evidence behind each letter and the fixes ranked. Run it before or after launch for a fast read; the deep nine-lane audit with rewrite fixes is suede-seo-audit.
"grade this page before I promote it"->/suede-visibility-graderAsk in plain words. The router reads the request and loads this lane; MCP agents find it with list_suede_skills.
claude code · sample session
$ /suede-visibility-grader yourpage.comfirst screen · CTA pull · proof · AI readabilitycited by answer engines: not yetVISIBILITY: C+ · FIX LIST RANKED
Use it before a page gets promoted, shared publicly, linked from a repo, added to an article, or pushed through Suedify. It pairs well with SEO/AEO/AI EO audits because it checks whether search, answer engines, Google result surfaces, Gemini result surfaces, and humans can all understand the same page.
For this repo, the grader should treat GitHub Pages as the actual public docs surface generated from the public repository, not as a separate marketing site.
First-screen clarityWhether the first viewport names the audience, outcome, proof, and action.
CTA pullPrimary action, secondary proof action, link target, button language, and why someone should click now.
Proof and trustScreenshots, commands, docs, manifests, source files, receipts, authorship, and safe evidence boundaries.
AI readabilityAnswer-ready copy, schema, sourceable claims, internal links, citation-friendly headings, and AI EO surface area.
Design signalHierarchy, spacing, type, image treatment, responsiveness, contrast, and whether the page looks intentional.
Output format
The grader starts with a simple non-coder explanation, then gives the usual breakdown, source/live status, screenshot evidence, viewport states, lane grades, grade caps, top fixes, CTA rewrites, verification notes, and a ship gate.
On major work, use the visibility grader as its own lane between builder and release verifier. Pair it with design review, A-F code grading, adversarial review, and final handoff so the page is checked from multiple angles before anyone claims it is ready.
Best prompts
Use $suede-visibility-grader to grade this page A-F for findability, first-screen clarity, CTA pull, proof, AI readability, rendered design signal, evidence, and promotion readiness.
Use $suede-visibility-grader on this GitHub Pages docs site. Treat it as the actual public repo docs surface, not a separate marketing site.
Use $suede-visibility-grader to check whether this page is ready for Google and Gemini result surfaces, then rewrite the primary and secondary CTAs.
Safety boundary
The A-F grade is a working review, not an audited business metric. The skill does not invent traffic, ranking, conversion, partner, legal, payout, or placement claims.
Keep Google and Gemini result receipts named for the surface they actually show.
Cue Suede
Feedback can happen mid-workflow or at the end. Say Cue Suede to ask for choices: change something, preserve what worked so the agent can mimic it later, or keep as-is by saying nothing.
Install this skill
It ships inside the Suede Creator Skills pack, so one install brings the whole pack and this skill with it. Claude Code takes two commands the first time and one after that; Codex installs the same pack natively.