Skip to content

seo: stop the description echoing the title, and clear the copy-voice defects - #15

Merged
JasonColapietro merged 1 commit into
mainfrom
seo/e2e-audit-20260905-low
Sep 5, 2026
Merged

seo: stop the description echoing the title, and clear the copy-voice defects#15
JasonColapietro merged 1 commit into
mainfrom
seo/e2e-audit-20260905-low

Conversation

@JasonColapietro

Copy link
Copy Markdown
Collaborator

Second pass on the 2026-09-05 end-to-end audit for dna.suedeai.ai. Four LOW findings,
all text or attribute level. No claim, number, name, date, or sentence subject changes.

dna.suedeai.ai#4 — home meta description restated the title verbatim.
src/lib/seo.ts homePageDescription() ended with "Suede DNA — signal chains, archived.",
a near character-for-character echo of the <title>. That clause now carries the sourcing
attribution the description did not otherwise state. The same function feeds the
CollectionPage JSON-LD description, so both surfaces move together.

dna.suedeai.ai#5 — em dashes as clause connectors on /about.
The audit named three, the file had seven in body prose. All are recast with a comma or a
period. The two remaining em dashes in the file are the About — Suede DNA OG and Twitter
title separators, which the finding exempts.

dna.suedeai.ai#6 — absolutes ("every", "nothing") in body copy.
grep -niE '\bevery\b|\bnothing\b' across the three named files now returns nothing.
The audit listed two instances on /about; there were three. The FAQ answer is also the
FAQPage acceptedAnswer.text, so the JSON-LD moves with the rendered text.

One deviation from the suggested rewrite: for the Clapton article the audit proposed
"Some details of that story may be embellished...", which asserts embellishment the
original left open. Kept the original's agnostic framing instead:
"Whether the details of that story are precise or slightly burnished...".

dna.suedeai.ai#7 — Lighthouse mobile label-content-name-mismatch, applied.
The header wordmark link carried aria-label="Suede home" while displaying SUEDE/DNA,
so its accessible name did not contain its visible label. It was the only element on the
served homepage with a name-from-content role and an overriding aria-label; the other
five sit on nav, aside, and role="group" containers, which the rule does not cover.
Dropping the override lets the name come from the visible text, which matches by
construction.

dna.suedeai.ai#7 — target-size, not fixed. Measured on the live page at 375x812:
16 targets under 24x24 CSS px, in three groups — the five header nav links (22.4x16.5,
inside a bar pinned to var(--header-main-h)), ten rig-card source links (325x16.5), and
the footer link (104.7x14). Every one is a .mono-label element whose height comes from
the type scale in src/styles/tokens.css. Raising them means changing spacing tokens and
layout across the header, the cards, and the footer, which is a design decision under the
visual overhaul spec rather than a mechanical fix, so it is left for that work.

Gates: pnpm typecheck, pnpm lint, pnpm test (111 passed), pnpm build all pass.

… defects

Four low-severity findings from the 2026-09-05 end-to-end audit, all of them
text or attribute level.

The home meta description spent its last sentence restating the title tag
almost character for character, so a reader who had already seen the title
learned nothing from it. That closing clause now carries the sourcing
attribution instead, which is a fact the description did not otherwise state.

The About page joined clauses with em dashes, and three pages carried the
absolutes the estate copy-voice rule bars. Each one is recast with a comma,
a period, or a scoped phrase. No claim, number, name, or date moves.

The header wordmark link labelled itself "Suede home" while displaying
SUEDE/DNA, so its accessible name did not contain its visible label and
Lighthouse scored label-content-name-mismatch at 0. Dropping the override
lets the name come from the visible text, which is what the rule asks for.

Findings: dna.suedeai.ai#4, #5, #6, #7 (label half).
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Sep 5, 2026

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review Completed 2026-09-05T21:26:00.215279Z 15aea7f PR opened
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@JasonColapietro
JasonColapietro merged commit f4c82cf into main Sep 5, 2026
1 check passed
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