Index: src/routes/posts.js
===================================================================
--- src/routes/posts.js	(revision 93b147d67da01570400474044f1ff3c35aeb6592)
+++ src/routes/posts.js	(revision 328d83725a5461bae2e5639ede37cbe4cf206b67)
@@ -19,4 +19,5 @@
 import { toWebp } from '../services/ImageWebpService.js';
 import ActivityPubService from '../services/ActivityPubService.js';
+import MusicMeta from '../services/MusicMeta.js';
 
 const __dirname = path.dirname(fileURLToPath(import.meta.url));
@@ -1014,4 +1015,5 @@
     siteAvatar,
     postHasPlayableAudio: ActivityPubService.hasPlayableAudio(post.content || '', site.id),
+    musicLd: MusicMeta.build((process.env.PUBLIC_BASE_URL || `${req.protocol}://${req.get('host')}`).replace(/\/+$/, ''), site, post),
     pageTitle: post.title + ' - ' + site.title,
     socialDescr: post.excerpt || '',
Index: src/services/MusicMeta.js
===================================================================
--- src/services/MusicMeta.js	(revision 328d83725a5461bae2e5639ede37cbe4cf206b67)
+++ src/services/MusicMeta.js	(revision 328d83725a5461bae2e5639ede37cbe4cf206b67)
@@ -0,0 +1,87 @@
+// Phase 1 of music federation: emit STANDARD schema.org MusicRecording / MusicAlbum
+// structured data for an audio post (Google rich results + any generic JSON-LD consumer).
+// Deliberately a real, existing web standard — NOT a Klonkt-invented field. The track
+// resolution here is reused by the (future) Funkwhale Audio/Library federation (Phase 2).
+import db from '../config/database.js';
+
+const COLS = 'title, album, duration, credit, license, cover_url, media_id';
+
+function isoDuration(sec) {
+  const n = parseInt(sec, 10);
+  if (!n || n < 0) return null;
+  return `PT${Math.floor(n / 60)}M${n % 60}S`; // ISO-8601 duration, e.g. PT3M20S
+}
+function absUrl(base, u) {
+  if (!u) return null;
+  return /^https?:/i.test(u) ? u : `${base}${u.startsWith('/') ? '' : '/'}${u}`;
+}
+
+// Resolve a post's [[track]]/[[album]]/[[playlist]] shortcodes to the HOSTED (playable)
+// tracks it references — only file-backed tracks (media_id), mirroring hasPlayableAudio.
+function resolveTracks(site, content) {
+  const tracks = [];
+  try {
+    for (const m of content.matchAll(/\[\[track:([A-Za-z0-9_-]+)\]\]/g)) {
+      const r = db.prepare(`SELECT ${COLS} FROM audio_tracks WHERE id = ?`).get(m[1]);
+      if (r && r.media_id) tracks.push(r);
+    }
+    for (const m of content.matchAll(/\[\[album:([^\]]+)\]\]/g)) {
+      for (const r of db.prepare(`SELECT ${COLS} FROM audio_tracks WHERE site_id = ? AND album = ? AND media_id IS NOT NULL ORDER BY rowid`).all(site.id, m[1].trim())) tracks.push(r);
+    }
+    for (const m of content.matchAll(/\[\[playlist:([A-Za-z0-9_-]+)\]\]/g)) {
+      for (const r of db.prepare(`SELECT t.title, t.album, t.duration, t.credit, t.license, t.cover_url, t.media_id FROM playlist_tracks pt JOIN audio_tracks t ON t.id = pt.track_id WHERE pt.playlist_id = ? AND t.media_id IS NOT NULL ORDER BY pt.position`).all(m[1])) tracks.push(r);
+    }
+  } catch { /* non-fatal */ }
+  return tracks;
+}
+
+// Build a schema.org MusicRecording (single track) or MusicAlbum (multiple) for a post,
+// or null when the post has no hosted audio. `url` points to the gated player page — the
+// anti-steal posture is preserved (no raw file URL is ever emitted).
+export function build(base, site, post) {
+  if (!post || !site || !post.content) return null;
+  if (!/\[\[(track|album|playlist):/i.test(post.content)) return null;
+  const b = (base || '').replace(/\/+$/, '');
+  const tracks = resolveTracks(site, post.content);
+  if (!tracks.length) return null;
+
+  const artist = {
+    '@type': 'MusicGroup',
+    name: site.title || site.slug,
+    url: `${b}/${site.is_primary ? '' : 'user/' + encodeURIComponent(site.slug)}`,
+  };
+  const postUrl = `${b}/${encodeURIComponent(post.slug)}`;
+  const recording = (t, withTop) => {
+    const o = { '@type': 'MusicRecording', name: t.title || post.title || 'Untitled' };
+    if (withTop) { o.byArtist = artist; o.url = postUrl; }
+    if (t.album) o.inAlbum = { '@type': 'MusicAlbum', name: t.album };
+    const d = isoDuration(t.duration); if (d) o.duration = d;
+    if (t.license) o.license = t.license;     // e.g. "CC BY 4.0" — Klonkt leads on this
+    if (t.credit) o.creditText = t.credit;
+    const cov = absUrl(b, t.cover_url) || absUrl(b, post.cover_image_url); if (cov) o.image = cov;
+    return o;
+  };
+
+  let ld;
+  if (tracks.length > 1) {
+    const albums = [...new Set(tracks.map((t) => t.album).filter(Boolean))];
+    ld = {
+      '@type': 'MusicAlbum',
+      name: albums.length === 1 ? albums[0] : (post.title || 'Album'),
+      byArtist: artist,
+      url: postUrl,
+      numTracks: tracks.length,
+      track: tracks.map((t) => recording(t, false)),
+    };
+    const cov = absUrl(b, post.cover_image_url) || absUrl(b, tracks[0].cover_url); if (cov) ld.image = cov;
+    const lic = tracks.find((t) => t.license); if (lic) ld.license = lic.license;
+  } else {
+    ld = recording(tracks[0], true);
+  }
+  ld['@context'] = 'https://schema.org';
+  const dp = post.published_at || post.created_at;
+  if (dp) ld.datePublished = new Date(dp).toISOString();
+  return ld;
+}
+
+export default { build };
Index: src/views/shell.ejs
===================================================================
--- src/views/shell.ejs	(revision 93b147d67da01570400474044f1ff3c35aeb6592)
+++ src/views/shell.ejs	(revision 328d83725a5461bae2e5639ede37cbe4cf206b67)
@@ -199,4 +199,9 @@
 <script type="application/ld+json"><%- JSON.stringify(_jsonLd) %></script>
 <% } %>
+<% if (typeof musicLd !== 'undefined' && musicLd) { %>
+<%# Music posts also carry standard schema.org MusicRecording/MusicAlbum data (Phase 1 of
+    music federation): real web standard, read by search engines + generic consumers. %>
+<script type="application/ld+json"><%- JSON.stringify(musicLd) %></script>
+<% } %>
 
 <!-- Self-hosted fonts (privacy-first) -->