feat(press): add link-only press kit for Meet Student Slovenia 2026

Unlisted MK press kit for media covering the 4 Sept 2026 event: fact
sheet, ready-to-quote boilerplate, key messages, downloadable event
photos + logos, and info@ contact. Reuses _press.css with new press-kit
blocks (one-line hero title, centered CTAs, tightened fact columns,
photo download grid).

Keep it link-only: add an UNLISTED mechanism to tools/seo_inject.py so
the page gets robots=noindex,nofollow on every run and is excluded from
sitemap.xml (and not exposed in robots.txt). Without this the pre-commit
SEO injector would make the page indexable and list it publicly.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

tools/seo_inject.py

@@ -29,6 +29,15 @@ LANGS = ["en", "mk", "si"]
 HREF = {"en": "en", "mk": "mk", "si": "sl"}
 OG_LOCALE = {"en": "en_GB", "mk": "mk_MK", "si": "sl_SI"}
 
+# Unlisted / link-only pages, keyed by (lang, path). These get a
+# `robots: noindex, nofollow` meta on every run and are kept OUT of sitemap.xml,
+# so they are reachable only by someone who has the direct URL (e.g. a press kit
+# shared privately with journalists). They are deliberately NOT added to
+# robots.txt — a Disallow line there would publicly advertise the secret path.
+UNLISTED = {
+    ("mk", "press/meet-student-slovenia-2026"),
+}
+
 # Section (first path segment) display names, for Article breadcrumbs.
 # Any 2-segment path "section/slug" whose section is here is treated as an article.
 SECTION_NAMES = {
@@ -458,6 +467,8 @@ def build_block(lang, path, content):
     ogimg = extract_og_image(content)
 
     lines = [START]
+    if (lang, path) in UNLISTED:
+        lines.append('    <meta name="robots" content="noindex, nofollow">')
     lines.extend(GA_LINES)
     if desc:
         lines.append(f'    <meta name="description" content="{html.escape(desc, quote=True)}">')
@@ -555,6 +566,8 @@ def write_sitemap():
     urls = []
     for L, fp in all_pages():
         path = page_path_of(fp, L)
+        if (L, path) in UNLISTED:
+            continue  # link-only page: keep it out of the public sitemap
         urls.append((L, path))
     # group by path so we can emit xhtml:link alternates
     from collections import defaultdict