Add Google Search Console verification

Author: Hilo-Hilo

.agents/skills/pre-production-checklist/SKILL.md

@@ -42,7 +42,7 @@ Run this gate whenever public routes, metadata, canonical-domain behavior, or th
 1. Treat `src/data/siteIdentity.json` and `src/data/sitePages.json` as the shared registries for public identity, canonical origin, indexable routes, and route-specific titles and descriptions. Compare them with the public routes in `src/App.tsx` and the main-page links in `public/llms.txt`.
 2. Run `npm test` and `npm run build`. Confirm every registered route produces `dist/<route>/index.html` with rendered page content rather than an empty React root.
 3. Serve `dist` with a plain static server and request every canonical route directly. Each registered route must return `200`; unknown and `/concepts` routes must remain non-indexable.
-4. Inspect each generated page's initial HTML. Confirm it has one unique title, description, self-referencing canonical URL, Open Graph URL, `index, follow` directive, and valid Organization/WebPage JSON-LD.
+4. Inspect each generated page's initial HTML. Confirm it has one unique title, description, self-referencing canonical URL, Open Graph URL, `index, follow` directive, Google Search Console verification tag, and valid Organization/WebPage JSON-LD.
 5. Confirm `dist/sitemap.xml` lists every registered production canonical exactly once and no staging, concept-review, or unknown URL. Confirm `dist/robots.txt` allows crawling and references the production sitemap.
 6. Run a staging build with `VITE_NOINDEX=true` and the staging `VITE_SITE_ORIGIN`. Confirm every generated page says `noindex, nofollow`, `dist/robots.txt` allows crawlers to fetch and observe that directive without advertising a sitemap, and no sitemap is published. On GitHub project Pages, remember that a robots file below the host root is not a host-wide crawl policy.
 7. After deployment, request the production routes, `/robots.txt`, and `/sitemap.xml` directly. In Google Search Console, submit the sitemap and use URL Inspection on the homepage and at least one inner page.

README.md

@@ -46,6 +46,7 @@ The validator reports unmatched responses, invalid public links, missing assets,
 ## Search and static routes
 
 - `src/data/siteIdentity.json` and `src/data/sitePages.json` are the shared registries for the public identity, indexable routes, page titles, and descriptions.
+- `src/data/siteIdentity.json` also stores the public Google Search Console verification token that the build emits as a metadata tag.
 - `npm run build` creates rendered HTML at each route directory, then generates and validates `dist/sitemap.xml` and `dist/robots.txt`.
 - `src/components/RouteMetadata.tsx` keeps metadata synchronized during client-side navigation.
 - Production builds are indexable by default. Staging sets `VITE_NOINDEX=true` and a staging `VITE_SITE_ORIGIN`, which marks every page `noindex, nofollow` and suppresses the sitemap. Its robots file allows fetching so crawlers can observe the page-level directive.

scripts/lib/site-seo.js

@@ -132,12 +132,16 @@ export const buildPageHtml = ({
   }
 
   html = html.replace(/\s*<meta\s+name="robots"[^>]*>/i, "");
+  html = html.replace(/\s*<meta\s+name="google-site-verification"[^>]*>/i, "");
   const robotsMeta = noindex
     ? '<meta name="robots" content="noindex, nofollow" />'
     : '<meta name="robots" content="index, follow" />';
+  const verificationMeta = siteIdentity.googleSiteVerification
+    ? `<meta name="google-site-verification" content="${escapeHtml(siteIdentity.googleSiteVerification)}" />`
+    : "";
   html = html.replace(
     "</head>",
-    `    ${robotsMeta}\n    <script id="site-structured-data" type="application/ld+json">${jsonForHtml(structuredData)}</script>\n  </head>`,
+    `    ${robotsMeta}\n    ${verificationMeta}\n    <script id="site-structured-data" type="application/ld+json">${jsonForHtml(structuredData)}</script>\n  </head>`,
   );
   html = replaceRequired(
     html,

scripts/lib/site-seo.test.js

@@ -11,6 +11,7 @@ const pages = [
 const siteIdentity = {
   name: "Example Organization",
   alternateName: "Example Org",
+  googleSiteVerification: "verification-token",
   logoPath: "/logo.svg",
   sameAs: ["https://social.example.com/example"],
 };
@@ -70,6 +71,10 @@ test("embeds rendered content and unique route metadata in the initial HTML", ()
   assert.match(html, /<main><h1>About us<\/h1><\/main>/);
   assert.match(html, /"@type":"WebPage"/);
   assert.match(html, /"@type":"EducationalOrganization"/);
+  assert.match(
+    html,
+    /<meta name="google-site-verification" content="verification-token" \/>/,
+  );
   assert.doesNotMatch(html, /name="robots" content="noindex/);
 });
 

scripts/validate-seo-build.mjs

@@ -38,6 +38,14 @@ for (const page of pages) {
   );
   assert.match(html, /"@type":"EducationalOrganization"/);
   assert.match(html, /"@type":"WebPage"/);
+  assert.equal(
+    occurrences(
+      html,
+      `<meta name="google-site-verification" content="${siteIdentity.googleSiteVerification}" />`,
+    ),
+    1,
+    `${page.path} must contain the Google Search Console verification token exactly once`,
+  );
   assert.match(
     html,
     noindex

src/data/siteIdentity.json

@@ -2,6 +2,7 @@
   "name": "Computational Biology at Berkeley",
   "alternateName": "CompBio at Berkeley",
   "productionOrigin": "https://compbioatberkeley.github.io",
+  "googleSiteVerification": "02GuXcaPxq10BM-UsozE-BtdcKgEJdfbfNYC1CUEYB0",
   "logoPath": "/comp-bio-logo-berkeley.svg",
   "sameAs": [
     "https://github.com/CompbioAtBerkeley",