Skip to content

Make an exported template a website rather than a deck - #373

Merged
singhharsh1708 merged 1 commit into
mainfrom
feat/exported-site-completeness
Sep 5, 2026
Merged

Make an exported template a website rather than a deck#373
singhharsh1708 merged 1 commit into
mainfrom
feat/exported-site-completeness

Conversation

@singhharsh1708

Copy link
Copy Markdown
Owner

You said the sites we hand people aren't proper. I exported one through the real button, unzipped it, served it and looked. Two things a visitor meets immediately were missing.

The one button on the page did nothing

Every template's call to action is an in-page link — #start, #signup, #book, sixteen distinct anchors across twenty-one templates. The exported page emitted no section ids at all:

ids in a real export: main  scroll-canvas  scroll-container  scroll-hint
CTA href:             #signup          <- nothing to land on

Sections now carry section-1, section-2, … and the five templates that open with a button point it at their closing section. Verified by clicking it on the served export:

clicking "Download TripVault" (#section-3)
scrollY 0 -> 3400   SCROLLED

The closing button is the owner's to fill in

A closing CTA has no in-page destination — only the owner knows whether it opens a signup page, a booking form or their inbox. So it is no longer shipped silently dead. The export names it:

One button needs a real link: Get it free. Set it in the section's CTA link field, then export again.

And the ZIP's README gained a section on setting it, covering https://, mailto: and tel:, plus how #section-N works. For tripvault the warning correctly reports one button, not two, because the opening one now resolves.

There was no footer

A finished site ended mid-scroll with nothing identifying its owner. It now closes with the site name, its description where there is one, and a copyright line — styled from --sc-ground, --sc-ink, --sc-muted and --sc-font-display, so it inherits the site's own look rather than a pasted-on one. Measured on the served page: 165px tall, correct ground colour, hairline border, © 2026 TripVault. All rights reserved.

Three claims the owner never made

was now
"Trusted by 4,200 teams" "Built for revenue teams"
"SOC 2 Type II, 99.99% uptime, and an audit log…" "Role-based access, a full audit log, and single sign-on…"
"4.9 stars · 2.1M downloads" "On iOS and Android"

A template's copy ships as the owner's own words, so it should describe the product, not invent traction. A test now scans all 21 for customer counts, star ratings and compliance certifications.

Unchanged

Lighthouse on the served export: 100 / 100 / 100 / 100 on mobile and desktop, LCP 1.3s mobile and 0.2s desktop — same as before the footer.

Eleven new tests, all failing on main. Suite 363 passed.

Still thin

Templates average 3.7 content sections and zero images. That is a content problem rather than a broken one, and a bigger piece of work — say the word and I will take it next.

Exported a template through the real button, unzipped it and served it. Two
things a visitor meets immediately were missing.

The one button on the page did nothing. Every template's call to action is an
in-page link - #start, #signup, #book, sixteen distinct anchors across
twenty-one templates - and the exported page emitted no section ids at all, so
none of them resolved. Sections now carry section-1, section-2 and so on, and
the five templates that open with a button point it at their closing section:
clicking it scrolls, verified at 0 to 3400px.

A closing call to action has no in-page destination, because only the owner
knows whether it should open a signup page, a booking form or their inbox. That
one is no longer shipped silently dead: the export names it - "One button needs
a real link: Get it free" - and the ZIP's README says how to set it, including
mailto: and tel:.

There was also no footer of any kind, so a finished site ended mid-scroll with
nothing identifying its owner. It now closes with the site's name, its
description where there is one, and a copyright line, styled from the theme's
own tokens.

Three pieces of template copy asserted things the owner had never said:
"Trusted by 4,200 teams", "SOC 2 Type II, 99.99% uptime", "4.9 stars, 2.1M
downloads". Replaced with claims about the product rather than its traction.

Lighthouse on the served export is unchanged at 100 for performance,
accessibility, best practices and SEO on both mobile and desktop.

Suite 363 passed; all eleven new tests fail on main.
@vercel

vercel Bot commented Sep 5, 2026

Copy link
Copy Markdown
Contributor

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated
scrollcraft Ready Ready Preview Sep 5, 2026 2:37pm UTC

@singhharsh1708
singhharsh1708 merged commit 0c6a21d into main Sep 5, 2026
3 checks passed
@singhharsh1708
singhharsh1708 deleted the feat/exported-site-completeness branch September 5, 2026 14:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant