Skip to content

docs(evidence): record live-deployment verification of finding b8f6046e942a #25

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
nish3451 merged 1 commit into from
Aug 8, 2026
Merged

docs(evidence): record live-deployment verification of finding b8f6046e942a #25

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 27 additions & 1 deletion docs/evidence/render-blocking-fonts-2026-08-08.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -113,4 +113,30 @@ has no external network dependency; the static guards above keep the real font
URL and the no-JS fallback honest. A blocking shape (a `<link rel="stylesheet">`
or `@import` returning on any of the six pages) fails CI: measured locally, a
reintroduced blocking link holds first paint to the delayed css2 response
(fcp 2780ms vs css2 end 2566ms) and the check fails with exit code 1.
(fcp 2780ms vs css2 end 2566ms) and the check fails with exit code 1.

### Live deployment verification (added 2026-08-09)

The limitation above — "The live tinystudio.io deployment was not measured here;
a deployed page could differ (CDN cache, different asset versions)" — is now
closed. On 2026-08-09 the deployed pages were measured in real Chromium
(Playwright 1.62.1, headless, unthrottled, no artificial CDN delay):

| Page | css2 renderBlockingStatus | FCP (ms) | css2 responseEnd (ms) | Render-blocking resources | Fonts load (Karla / Fraunces) | Promoted sheet applied |
|---|---|---|---|---|---|---|
| index.html (home) | non-blocking | 652 | 460 | same-origin index.css only | yes / yes | yes |
| audit.html | non-blocking | 624 | 536 | same-origin audit.css, shared.css | yes / yes | yes |
| agents.html | non-blocking | 880 | 558 | same-origin agents.css, shared.css | yes / yes | yes |
| pricing.html | non-blocking | 720 | 531 | same-origin shared.css, pricing.css | yes / yes | yes |
| specimen.html | non-blocking | 796 | 623 | same-origin specimen.css, shared.css | yes / yes | yes |
| brief-requested.html | non-blocking | 668 | 527 | same-origin shared.css, brief-requested.css | yes / yes | yes |

The only render-blocking resources on any live page are the site's own
same-origin stylesheets — the exact allowance the CI check
(`scripts/check-render-blocking.mjs`) asserts. No render-blocking scripts and no
render-blocking external resources were observed, including on the homepage the
original dogfood run flagged. The Google Fonts css2 stylesheet is fetched
non-blocking on all six live pages, first paint does not wait for it, and fonts

@chatgpt-codex-connector chatgpt-codex-connector Bot Aug 8, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Verify live paint timing before claiming it does not wait

The live run is explicitly unthrottled with no artificial CDN delay, and every row shows FCP occurring after the css2 response ended, so these measurements do not demonstrate that first paint would proceed while css2 remained pending. Under the receipt's own fcp >= css2End timing check, all six rows would be classified as having waited; proving this causal claim requires a delayed live response as in the earlier timing pass, otherwise the claim should be limited to the observed renderBlockingStatus.

Useful? React with 👍 / 👎.

still load and apply under the production CSP. This closes dogfood finding
b8f6046e942a ("Render-blocking resources on home") against the deployed site;
the code-side fix and the CI enforcement were already merged as PRs #20 and #23.
Loading