Skip to content

armgit / kiln

Public Imported from GitHub
0

Build and revise procedural 3D assets with your coding agent. Local MCP server, CLI and TypeScript engine with rendered review, editable source and GLB export.

Keep raw data files out of search and add a footer contact link (#332)

Matthew Kissingerbe55d83ec646
diff --git a/docs/site-indexing.md b/docs/site-indexing.md
index d3d2fa21..389509ef 100644
--- a/docs/site-indexing.md
+++ b/docs/site-indexing.md
@@ -103,3 +103,30 @@ sits under a retired directory, that each target directory has files (in a packs
build; the reduced `KILN_SITE_PACKS=0` build has no scene packs), and that no target
is itself redirected. Asset JSON and GLB 404s and the retired `play.` hosts
are left alone.
+
+## Raw data files — November 5, 2026
+
+Search Console showed Google crawling raw files as documents, for example
+`/asset-demos/**/manifest.json`, `/asset-demos/**/*.kiln-metadata.json` and
+`/scene-packs/**/THIRD-PARTY-NOTICES.txt`. `_headers` now sends `X-Robots-Tag: noindex`
+on `.json`, `.txt`, `.md` and `SHA256SUMS` under `/asset-demos/` and `/scene-packs/`, and on the
+same kind of file elsewhere: `/assets/*.json`, `/scene-runtime/*.json`, `/sources/*.md`, `/fonts/*.txt`,
+licence and lock files beside the raw data, and the root build receipts.
+The header does not stop a fetch: scenes and agents load these files as before. `/llms.txt`
+is deliberately left indexable so agents and AI tools can find it. `robots.txt` stays allow-all, because a crawler cannot see a noindex on a
+blocked URL. `validate-static` fails a build where a raw sample loses the header, where a
+rule reaches a built `.html` file, a sitemap URL, `llms.txt`, the media a page shows or
+`.well-known`, or where `robots.txt` blocks a raw file. GLBs, images and the `scene-source`
+code files are unchanged. To check production after a deploy:
+
+```sh
+curl -sI https://kilnstudio.tools/scene-packs/golden-gate/g9-code6/SHA256SUMS | grep -i x-robots # noindex
+curl -sI https://kilnstudio.tools/assets/abyssal-surveyor.poster.json | grep -i x-robots # noindex
+curl -sI https://kilnstudio.tools/llms.txt | grep -i x-robots # nothing
+curl -sI https://kilnstudio.tools/ | grep -i x-robots # nothing
+curl -s https://kilnstudio.tools/ | grep -c 'cdn-cgi/l/email-protection' # 0
+```
+
+The footer's Contact link sits inside `<!--email_off-->`, because Cloudflare's Email Address
+Obfuscation is on for this zone outside `/docs/`. It would otherwise rewrite the link to a
+`/cdn-cgi/` URL that answers 404 to crawlers and to visitors without JavaScript.
diff --git a/site/public/_headers b/site/public/_headers
index 425ea5db..e82a5b53 100644
--- a/site/public/_headers
+++ b/site/public/_headers
@@ -26,6 +26,70 @@ https://kilnstudio.pages.dev/*
https://:version.kilnstudio.pages.dev/*
X-Robots-Tag: noindex
+# Raw data the scenes and gallery fetch: Search Console showed Google crawling these as documents
+# (manifest.json, *.kiln-metadata.json, THIRD-PARTY-NOTICES.txt). X-Robots-Tag keeps a file out of the index and still
+# lets anything fetch it, so the demos keep working. Never list these in robots.txt: a crawler cannot see
+# the noindex on a blocked URL. One splat per rule and it crosses "/" (see headersFor), so "*.json" covers any depth.
+# HTML is deliberately absent, and validate-static fails a build where one of these reaches a page or a sitemap URL.
+/asset-demos/*.json
+ X-Robots-Tag: noindex
+
+/asset-demos/*.txt
+ X-Robots-Tag: noindex
+
+/asset-demos/*.md
+ X-Robots-Tag: noindex
+
+/asset-demos/*/SHA256SUMS
+ X-Robots-Tag: noindex
+
+/scene-packs/*.json
+ X-Robots-Tag: noindex
+
+/scene-packs/*.txt
+ X-Robots-Tag: noindex
+
+/scene-packs/*.md
+ X-Robots-Tag: noindex
+
+/scene-packs/*/SHA256SUMS
+ X-Robots-Tag: noindex
+
+# The same kind of file elsewhere: per-asset JSON the gallery fetches, runtime data, source notes, licence texts and
+# build receipts. llms.txt, robots.txt, the sitemaps and .well-known stay indexable.
+/asset-demos/*/LICENSE
+ X-Robots-Tag: noindex
+
+/asset-demos/*.lock
+ X-Robots-Tag: noindex
+
+/asset-demos/*/.gitignore
+ X-Robots-Tag: noindex
+
+/scene-packs/*/LICENSE
+ X-Robots-Tag: noindex
+
+/assets/*.json
+ X-Robots-Tag: noindex
+
+/scene-runtime/*.json
+ X-Robots-Tag: noindex
+
+/sources/*.md
+ X-Robots-Tag: noindex
+
+/fonts/*.txt
+ X-Robots-Tag: noindex
+
+/build-info.json
+ X-Robots-Tag: noindex
+
+/artifact-files.json
+ X-Robots-Tag: noindex
+
+/source-files.json
+ X-Robots-Tag: noindex
+
# Build output named by content hash: a changed file gets a new name, so these never need revalidating.
/_astro/*
! Cache-Control
diff --git a/site/scripts/static-validation-core.mjs b/site/scripts/static-validation-core.mjs
index 41fbfb5d..78facf3a 100644
--- a/site/scripts/static-validation-core.mjs
+++ b/site/scripts/static-validation-core.mjs
@@ -158,7 +158,12 @@ export function ownerReviewErrors(html) {
}
/** The copy rules every page answers to: numbers not glued to words, the site's spelling of licence, owner status. */
-export const copyErrors = (html) => [...gluedNumberErrors(html), ...licenceSpellingErrors(html), ...ownerReviewErrors(html)];
+/** Cloudflare's Email Address Obfuscation rewrites a mailto link to a /cdn-cgi/ URL that 404s without JavaScript. */
+export const emailLinkErrors = (html) => (/href="mailto:/i.test(html.replace(/<!--email_off-->[\s\S]*?<!--\/email_off-->/g, ''))
+ ? ['A mailto link outside <!--email_off--> would be rewritten by Cloudflare to a /cdn-cgi/ URL']
+ : []);
+
+export const copyErrors = (html) => [...emailLinkErrors(html), ...gluedNumberErrors(html), ...licenceSpellingErrors(html), ...ownerReviewErrors(html)];
/** Meta descriptions over this length are cut in search and share previews (content review finding 13). */
export const DESCRIPTION_WARNING_LENGTH = 160;
@@ -297,6 +302,10 @@ export function discoveryErrors({ sitemapUrls, robots, sitemapIndex }) {
const value = (name) => directives.filter((line) => line.toLowerCase().startsWith(`${name}:`)).map((line) => line.slice(name.length + 1).trim());
if (!value('sitemap').includes(`${ORIGIN}/sitemap.xml`)) add('/robots.txt', `robots.txt does not name ${ORIGIN}/sitemap.xml`);
if (value('disallow').includes('/')) add('/robots.txt', 'robots.txt blocks the whole site with "Disallow: /"');
+ // A crawler cannot see the noindex header on a URL robots.txt keeps it from fetching.
+ for (const rule of new Set(value('disallow').filter((rule) => rule && rule !== '/'))) {
+ if (RAW_NOINDEX_SAMPLES.some((path) => path.startsWith(rule))) add('/robots.txt', `robots.txt blocks ${rule}, so a crawler can never see its noindex header`);
+ }
}
return errors;
}
@@ -382,6 +391,47 @@ export function headersFor(rules, pathname, host = new URL(ORIGIN).host) {
return Object.fromEntries(headers);
}
+/** Raw files that must stay out of the index on the production host (site/public/_headers), one of each kind. */
+export const RAW_NOINDEX_SAMPLES = [
+ '/asset-demos/cabin-woods-psx-horror/metadata/rock/manifest.json',
+ '/asset-demos/cabin-woods-psx-horror/models/rock.kiln-metadata.json',
+ '/scene-packs/golden-gate/g9-code6/THIRD-PARTY-NOTICES.txt',
+ '/scene-packs/golden-gate/g9-code6/licenses/FEDERAL-DATA.txt',
+ '/scene-packs/golden-gate/g9-code6/pack.json',
+ '/scene-packs/golden-gate/g9-code6/SHA256SUMS',
+ '/scene-packs/golden-gate/g9-code6/README.md',
+ '/scene-packs/troy/troy-20261005-07/web/vendor/three/LICENSE',
+ '/asset-demos/cabin-woods-psx-horror/THIRD-PARTY-NOTICES.txt',
+ '/asset-demos/cabin-woods-psx-horror/scene-source/README.md',
+ '/asset-demos/cabin-woods-psx-horror/scene-source/LICENSE',
+ '/asset-demos/cabin-woods-psx-horror/scene-source/bun.lock',
+ '/asset-demos/cabin-woods-psx-horror/scene-source/.gitignore',
+ '/asset-demos/cabin-woods-psx-horror/SHA256SUMS',
+ '/assets/abyssal-surveyor.poster.json',
+ '/assets/abyssal-surveyor.kiln-metadata.json',
+ '/scene-runtime/farm/runtime.json',
+ '/sources/golden-gate-bridge/REFERENCE.md',
+ '/fonts/Archivo-OFL.txt',
+ '/build-info.json',
+ '/artifact-files.json',
+ '/source-files.json',
+];
+
+/** What search and agents read directly, and the media a page shows. A rule that reached one of these would cost traffic. */
+export const KEEP_INDEXABLE_SAMPLES = [
+ '/llms.txt',
+ '/robots.txt',
+ '/sitemap.xml',
+ '/sitemap-index.xml',
+ '/.well-known/agent-skills/index.json',
+ '/assets/abyssal-surveyor.glb',
+ '/assets/abyssal-surveyor.png',
+ '/assets/abyssal-surveyor.webp',
+ '/media/hero.png',
+ '/media/hero.webp',
+ '/asset-demos/cabin-woods-psx-horror/models/rock.glb',
+];
+
/** A script element's content hash as a CSP source expression. */
export const scriptHashSource = (text) => `'sha256-${createHash('sha256').update(text, 'utf8').digest('base64')}'`;
@@ -397,7 +447,7 @@ const ASTRO_TRANSFORMED_IMAGE = /^_astro\/[^/]+\.[A-Za-z0-9_-]{8}_[A-Za-z0-9]{1,
* carry. HTML revalidates on every visit; only files with a content hash in their name are cached as immutable, and
* every file under `/_astro/` is.
*/
-export function headersErrors({ text, inlineScripts = [], files = [], sealedFiles = [] }) {
+export function headersErrors({ text, inlineScripts = [], files = [], sealedFiles = [], sitemapPaths = [] }) {
const errors = [];
const sealed = new Set(sealedFiles);
let rules;
@@ -414,6 +464,12 @@ export function headersErrors({ text, inlineScripts = [], files = [], sealedFile
if (!policy) errors.push('Pages carry no Content-Security-Policy-Report-Only');
// The Pages project also answers on its pages.dev hosts with the same bytes; only the production host is indexed.
if (/noindex/i.test(page['x-robots-tag'] ?? '')) errors.push(`The production host ${new URL(ORIGIN).host} must not be served X-Robots-Tag: noindex`);
+ // Raw data is noindexed by header; a rule that reached a page or a sitemap URL would drop it from search.
+ const noindexed = (path) => /noindex/i.test(headersFor(rules, path)['x-robots-tag'] ?? '');
+ for (const path of RAW_NOINDEX_SAMPLES) if (!noindexed(path)) errors.push(`${path} must be served X-Robots-Tag: noindex`);
+ for (const path of KEEP_INDEXABLE_SAMPLES) if (noindexed(path)) errors.push(`${path} must not be served X-Robots-Tag: noindex`);
+ for (const file of files) if (/\.html$/.test(file) && noindexed(`/${file}`)) errors.push(`The page /${file} must not be served X-Robots-Tag: noindex`);
+ for (const path of sitemapPaths) if (noindexed(path)) errors.push(`The sitemap URL ${path} must not be served X-Robots-Tag: noindex`);
for (const host of PAGES_DEV_HOSTS) {
for (const path of ['/', '/docs/', '/robots.txt']) {
if (!/noindex/i.test(headersFor(rules, path, host)['x-robots-tag'] ?? '')) errors.push(`https://${host}${path} must be served X-Robots-Tag: noindex`);
diff --git a/site/scripts/static-validation.test.ts b/site/scripts/static-validation.test.ts
index bc419846..9541ffb4 100644
--- a/site/scripts/static-validation.test.ts
+++ b/site/scripts/static-validation.test.ts
@@ -1,5 +1,5 @@
import { describe, expect, test } from 'bun:test';
-import { ARCHIVE_INDEX_ROUTE, archiveIndexingErrors, copyErrors, discoveryErrors, embeddedDocumentErrors, foundryFloorErrors, headersErrors, headersFor, inspectHtml, parseHeaderRules, scriptHashSource, isArchiveItemRoute, isEmbeddedDocument, licenceSpellingErrors, ownerReviewErrors, parseRedirectRules, redirectsErrors, resolveInternalLink, retiredNameErrors, routeForFile, socialMetadataErrors } from './static-validation-core.mjs';
+import { ARCHIVE_INDEX_ROUTE, archiveIndexingErrors, copyErrors, discoveryErrors, embeddedDocumentErrors, foundryFloorErrors, emailLinkErrors, headersErrors, headersFor, inspectHtml, parseHeaderRules, scriptHashSource, isArchiveItemRoute, isEmbeddedDocument, licenceSpellingErrors, ownerReviewErrors, parseRedirectRules, redirectsErrors, resolveInternalLink, retiredNameErrors, routeForFile, socialMetadataErrors } from './static-validation-core.mjs';
import foundryFloor from '../src/data/foundry-floor.json';
import { readFileSync } from 'node:fs';
test('static routes use directory URLs while 404 keeps its HTML route', () => {
@@ -252,6 +252,8 @@ describe('discovery files', () => {
expect(check({ robots: 'User-agent: *\nAllow: /\n' })).toEqual(['/robots.txt: robots.txt does not name https://kilnstudio.tools/sitemap.xml']);
expect(check({ robots: 'User-agent: *\nDisallow: /\n\nSitemap: https://kilnstudio.tools/sitemap.xml\n' })).toEqual(['/robots.txt: robots.txt blocks the whole site with "Disallow: /"']);
expect(check({ robots: '' })).toEqual(['/robots.txt: robots.txt is missing or empty']);
+ // A rule that blocks raw data would hide its noindex header from crawlers.
+ expect(check({ robots: 'User-agent: *\nDisallow: /asset-demos/\n\nSitemap: https://kilnstudio.tools/sitemap.xml\n' })).toEqual(['/robots.txt: robots.txt blocks /asset-demos/, so a crawler can never see its noindex header']);
// A rule for one path is not the whole site, and an empty Disallow allows everything.
expect(check({ robots: 'User-agent: *\nDisallow: /private/\nDisallow:\n\nSitemap: https://kilnstudio.tools/sitemap.xml\n' })).toEqual([]);
});
@@ -385,12 +387,69 @@ https://kilnstudio.pages.dev/*
https://:version.kilnstudio.pages.dev/*
X-Robots-Tag: noindex
+
+/asset-demos/*.json
+ X-Robots-Tag: noindex
+
+/scene-packs/*.json
+ X-Robots-Tag: noindex
+
+/scene-packs/*.txt
+ X-Robots-Tag: noindex
+
+/scene-packs/*/SHA256SUMS
+ X-Robots-Tag: noindex
+
+/asset-demos/*.txt
+ X-Robots-Tag: noindex
+
+/asset-demos/*.md
+ X-Robots-Tag: noindex
+
+/asset-demos/*/SHA256SUMS
+ X-Robots-Tag: noindex
+
+/scene-packs/*.md
+ X-Robots-Tag: noindex
+
+/asset-demos/*/LICENSE
+ X-Robots-Tag: noindex
+
+/asset-demos/*.lock
+ X-Robots-Tag: noindex
+
+/asset-demos/*/.gitignore
+ X-Robots-Tag: noindex
+
+/scene-packs/*/LICENSE
+ X-Robots-Tag: noindex
+
+/assets/*.json
+ X-Robots-Tag: noindex
+
+/scene-runtime/*.json
+ X-Robots-Tag: noindex
+
+/sources/*.md
+ X-Robots-Tag: noindex
+
+/fonts/*.txt
+ X-Robots-Tag: noindex
+
+/build-info.json
+ X-Robots-Tag: noindex
+
+/artifact-files.json
+ X-Robots-Tag: noindex
+
+/source-files.json
+ X-Robots-Tag: noindex
${extra}`;
const good = file(`default-src 'self'; script-src 'self' ${scriptHashSource(script)}`);
test('rules apply in order, "!" removes what an earlier rule set, placeholders match one segment', () => {
const rules = parseHeaderRules(good);
- expect(rules.map((rule) => `${rule.host ?? ''}${rule.path}`)).toEqual(['/*', '/_astro/*', '/scene-runtime/:scene/*.js', 'kilnstudio.pages.dev/*', ':version.kilnstudio.pages.dev/*']);
+ expect(rules.map((rule) => `${rule.host ?? ''}${rule.path}`)).toEqual(['/*', '/_astro/*', '/scene-runtime/:scene/*.js', 'kilnstudio.pages.dev/*', ':version.kilnstudio.pages.dev/*', '/asset-demos/*.json', '/scene-packs/*.json', '/scene-packs/*.txt', '/scene-packs/*/SHA256SUMS', '/asset-demos/*.txt', '/asset-demos/*.md', '/asset-demos/*/SHA256SUMS', '/scene-packs/*.md', '/asset-demos/*/LICENSE', '/asset-demos/*.lock', '/asset-demos/*/.gitignore', '/scene-packs/*/LICENSE', '/assets/*.json', '/scene-runtime/*.json', '/sources/*.md', '/fonts/*.txt', '/build-info.json', '/artifact-files.json', '/source-files.json']);
expect(headersFor(rules, '/')['cache-control']).toBe('public, max-age=0, must-revalidate');
expect(headersFor(rules, '/_astro/viewer.GoKKfp1n.js')['cache-control']).toBe('public, max-age=31536000, immutable');
expect(headersFor(rules, '/scene-runtime/golden-gate/index-CA13LcNQ.js')['cache-control']).toBe('public, max-age=31536000, immutable');
@@ -447,7 +506,49 @@ ${extra}`;
const open = good.replace('\nhttps://kilnstudio.pages.dev/*\n X-Robots-Tag: noindex\n', '\n');
expect(headersErrors({ text: open, inlineScripts: [script] })).toEqual(['/', '/docs/', '/robots.txt'].map((path) => `https://kilnstudio.pages.dev${path} must be served X-Robots-Tag: noindex`));
const everywhere = `${good}\n/*\n X-Robots-Tag: noindex\n`;
- expect(headersErrors({ text: everywhere, inlineScripts: [script] })).toEqual(['The production host kilnstudio.tools must not be served X-Robots-Tag: noindex']);
+ expect(headersErrors({ text: everywhere, inlineScripts: [script] })).toContain('The production host kilnstudio.tools must not be served X-Robots-Tag: noindex');
+ });
+
+ test('raw data is noindexed by header on production, never a page, a sitemap URL or a robots.txt block', () => {
+ const published = readFileSync(new URL('../public/_headers', import.meta.url), 'utf8');
+ const rules = parseHeaderRules(published);
+ const robots = (path: string) => headersFor(rules, path)['x-robots-tag'];
+ for (const path of ['/asset-demos/a/metadata/rock/manifest.json', '/asset-demos/a/b/c/rock.kiln-metadata.json', '/asset-demos/a/THIRD-PARTY-NOTICES.txt', '/scene-packs/a/b/THIRD-PARTY-NOTICES.txt', '/scene-packs/a/b/licenses/vehicles/suv.ASSET-LICENSE.txt', '/scene-packs/a/b/data/layout.json', '/scene-packs/a/b/SHA256SUMS', '/assets/a.poster.json', '/assets/a.kiln-metadata.json', '/scene-runtime/a/runtime.json', '/sources/a/REFERENCE.md', '/fonts/A-OFL.txt', '/build-info.json']) {
+ expect(robots(path)).toBe('noindex');
+ }
+ for (const path of ['/', '/llms.txt', '/docs/', '/docs/install/', '/gallery/', '/gallery/some-asset/', '/packs/farm/', '/scenes/farm/', '/robots.txt', '/sitemap.xml', '/asset-demos/a/web/game.html', '/asset-demos/a/scene-source/index.html', '/asset-demos/a/models/rock.glb', '/scene-packs/troy/t/web/', '/scene-packs/troy/t/web/qualification/a.html', '/scene-packs/a/b/models/x.glb', '/assets/a.glb', '/assets/a.webp', '/fonts/a.woff2', '/.well-known/agent-skills/index.json', '/sitemap-index.xml']) {
+ expect(robots(path)).toBeUndefined();
+ }
+ expect(headersErrors({ text: published.replace(/sha256-[^']+/g, 'x') }).filter((error) => /X-Robots-Tag/.test(error))).toEqual([]);
+ });
+
+ test('every kind of raw file is guarded, and a rule that reaches what must stay indexable is reported', () => {
+ for (const rule of ['/asset-demos/*.md', '/scene-packs/*.md', '/asset-demos/*/SHA256SUMS', '/asset-demos/*.txt', '/assets/*.json', '/fonts/*.txt', '/build-info.json']) {
+ const removed = good.replace(`${rule}\n X-Robots-Tag: noindex\n`, '');
+ expect(removed).not.toBe(good);
+ expect(headersErrors({ text: removed, inlineScripts: [script] }).filter((error) => /must be served X-Robots-Tag/.test(error)).length).toBeGreaterThan(0);
+ }
+ for (const rule of ['/*.png', '/*.webp', '/media/*', '/llms.txt', '/*.txt', '/sitemap.xml', '/*.glb', '/.well-known/*']) {
+ const errors = headersErrors({ text: `${good}\n${rule}\n X-Robots-Tag: noindex\n`, inlineScripts: [script] });
+ expect(errors.filter((error) => /must not be served X-Robots-Tag/.test(error)).length).toBeGreaterThan(0);
+ }
+ });
+
+ test('a mailto link is only allowed inside email_off, where Cloudflare leaves it alone', () => {
+ expect(emailLinkErrors('<a href="https://example.com/">Site</a>')).toEqual([]);
+ expect(emailLinkErrors('<!--email_off--><a href="mailto:[email protected]">Contact</a><!--/email_off-->')).toEqual([]);
+ expect(emailLinkErrors('<a href="mailto:[email protected]">Contact</a>')).toHaveLength(1);
+ expect(emailLinkErrors('<!--email_off--><a href="mailto:[email protected]">A</a><!--/email_off--><a href="mailto:[email protected]">B</a>')).toHaveLength(1);
+ });
+
+ test('a raw data rule that is missing, or reaches a page or a sitemap URL, is reported', () => {
+ expect(headersErrors({ text: good.replace('/asset-demos/*.json\n X-Robots-Tag: noindex\n', ''), inlineScripts: [script] })).toEqual(['/asset-demos/cabin-woods-psx-horror/metadata/rock/manifest.json must be served X-Robots-Tag: noindex', '/asset-demos/cabin-woods-psx-horror/models/rock.kiln-metadata.json must be served X-Robots-Tag: noindex']);
+ const wide = `${good}\n/asset-demos/*\n X-Robots-Tag: noindex\n`;
+ expect(headersErrors({ text: wide, inlineScripts: [script], files: ['asset-demos/a/web/game.html', 'asset-demos/a/models/rock.glb'], sitemapPaths: ['/asset-demos/a/', '/docs/'] }).filter((error) => /must not/.test(error))).toEqual([
+ '/asset-demos/cabin-woods-psx-horror/models/rock.glb must not be served X-Robots-Tag: noindex',
+ 'The page /asset-demos/a/web/game.html must not be served X-Robots-Tag: noindex',
+ 'The sitemap URL /asset-demos/a/ must not be served X-Robots-Tag: noindex',
+ ]);
});
test('inline scripts are found in pages; JSON-LD data blocks are not scripts to a CSP', () => {
diff --git a/site/scripts/validate-static.mjs b/site/scripts/validate-static.mjs
index a46ccf88..b6140908 100644
--- a/site/scripts/validate-static.mjs
+++ b/site/scripts/validate-static.mjs
@@ -214,7 +214,7 @@ let headersText = '';
try { headersText = await readFile(resolve(root, '_headers'), 'utf8'); } catch { add('/_headers', 'headers', 'dist/_headers is missing (site/public/_headers)'); }
let sealedFiles=[];
try {sealedFiles=(await sealedTroyCache(root)).files;} catch(error) {add('/scene-packs/troy/','headers',error.message);}
-const headerProblems = headersText ? headersErrors({ text: headersText, inlineScripts, files: distFiles, sealedFiles }) : [];
+const headerProblems = headersText ? headersErrors({ text: headersText, inlineScripts, files: distFiles, sealedFiles, sitemapPaths: sitemapUrls.map((url) => new URL(url).pathname) }) : [];
for (const message of headerProblems) add('/_headers', 'headers', message);
const headerRules = headersText ? parseHeaderRules(headersText) : [];
const immutableFiles = distFiles.filter((file) => /immutable/.test(headersFor(headerRules, `/${file}`)['cache-control'] ?? ''));
diff --git a/site/src/layouts/BaseLayout.astro b/site/src/layouts/BaseLayout.astro
index 63322250..66f2d25c 100644
--- a/site/src/layouts/BaseLayout.astro
+++ b/site/src/layouts/BaseLayout.astro
@@ -258,6 +258,8 @@ const structured = [
<a class="text-link" href={DISCORD}>
Discord community
</a>
+ {/* email_off: Cloudflare would otherwise rewrite this to a /cdn-cgi/ URL that 404s without JavaScript. */}
+ <!--email_off--><a class="text-link" href="mailto:[email protected]">Contact</a><!--/email_off-->
<a class="text-link" href={`${REPO}/blob/main/LICENSE`}>
Kiln licence: MIT
</a>