diff --git a/Plugin/unraid/pages/monitor.php b/Plugin/unraid/pages/monitor.php
index 5b9a99c..9e8c25c 100644
--- a/Plugin/unraid/pages/monitor.php
+++ b/Plugin/unraid/pages/monitor.php
@@ -86,7 +86,12 @@
 // ═══════════════════════════════════════════════════════════════════════════════════════════════
 require_once dirname(__DIR__) . '/include/monitor.php';
 require_once dirname(__DIR__) . '/include/ai_chat.php';
+require_once dirname(__DIR__) . '/include/docs.php';
 if (vv_ai_ui_on()) vv_ai_chat_assets();
+
+// Live values for the `$VAR` markers in pages/readme/monitor-readme.md. Same arrangement the
+// Scheduler tab uses: the doc names a conf variable and the reader sees this host's value.
+$_vv_doc_vars = array_merge(vv_conf_vars(), ['SCRIPTS_DIR' => SCRIPTS_DIR]);
 ?>
 <style>
 @keyframes vvRsPulse {
@@ -348,6 +353,33 @@ if (vv_ai_ui_on()) vv_ai_chat_assets();
 
 </div>
 
+<!-- ── How to read this page ────────────────────────────────────────────────────
+     Rendered from pages/readme/monitor-readme.md, not maintained here. That file is also what
+     the AI tab retrieves, so the panel you read and the answer the assistant gives are the same
+     text and cannot drift.
+
+     Below the grid and collapsed by default: this page is watched, not read, and help that
+     pushed the cards down would be in the way every time you opened the tab for the reason you
+     usually open it. Two views of one file — the brief list is one line per card, "More info"
+     swaps in the full document for the parts that need explaining. -->
+<div class="vv-sug-block" data-save-key="monitor-howto" style="margin-top:12px;">
+  <div class="vv-sug-header" onclick="vvMonToggleHelp(this)">
+    <span class="vv-sug-chevron">▸</span>
+    <span class="vv-sug-title">How to read this page</span>
+    <button class="vv-more-btn" id="vv-mon-howto-more" type="button"
+            onclick="event.stopPropagation(); vvMonToggleMore(this)">More info</button>
+  </div>
+  <div class="vv-sug-body vv-info-body" style="display:none">
+    <ul class="vv-info-cols" id="vv-mon-howto-brief">
+      <?= vv_docs_brief('Plugin/unraid/pages/readme/monitor-readme.md', $_vv_doc_vars,
+                        fn($h) => is_string($h) && str_starts_with($h, 'Reference —')) ?>
+    </ul>
+    <div class="vv-doc" id="vv-mon-howto-full" style="display:none">
+      <?= vv_docs_render('Plugin/unraid/pages/readme/monitor-readme.md', $_vv_doc_vars) ?>
+    </div>
+  </div>
+</div>
+
 <script>
 // ── Shared helpers ────────────────────────────────────────────────────────────
 
@@ -2601,6 +2633,42 @@ document.addEventListener('click', () => {
 // Removed rather than left as an unreachable handler for the platform's three most destructive
 // operations. The endpoint stays; see its header for why it is kept unwired.
 
+// ── Help panel ───────────────────────────────────────────────────────────────
+// Collapse state and depth are both remembered. Which one you want is a preference, not a
+// per-visit decision — and on a page you leave open, re-collapsing help you deliberately opened
+// would be its own small annoyance.
+function vvMonToggleHelp(header) {
+  const body    = header.nextElementSibling;
+  const chevron = header.querySelector('.vv-sug-chevron');
+  const open    = body.style.display !== 'none';
+  body.style.display  = open ? 'none' : '';
+  chevron.textContent = open ? '▸' : '▾';
+  localStorage.setItem('vv-sug-monitor-howto', open ? '0' : '1');
+}
+
+function vvMonToggleMore(btn) {
+  const brief = document.getElementById('vv-mon-howto-brief');
+  const full  = document.getElementById('vv-mon-howto-full');
+  if (!brief || !full) return;
+  const showFull = full.style.display === 'none';
+  full.style.display  = showFull ? '' : 'none';
+  brief.style.display = showFull ? 'none' : '';
+  btn.textContent     = showFull ? 'Less' : 'More info';
+  btn.classList.toggle('vv-more-on', showFull);
+  localStorage.setItem('vv-mon-howto-more', showFull ? '1' : '0');
+}
+
+(function vvMonRestoreHelp() {
+  if (localStorage.getItem('vv-sug-monitor-howto') === '1') {
+    const h = document.querySelector('[data-save-key="monitor-howto"] .vv-sug-header');
+    if (h) vvMonToggleHelp(h);
+  }
+  if (localStorage.getItem('vv-mon-howto-more') === '1') {
+    const b = document.getElementById('vv-mon-howto-more');
+    if (b) vvMonToggleMore(b);
+  }
+})();
+
 // ── AI row ───────────────────────────────────────────────────────────────────
 // Constructed only when the row rendered. The card markup is behind vv_ai_ui_on(), so on any
 // other host these ids do not exist and the factories are never called — the row is absent
diff --git a/Plugin/unraid/pages/readme/monitor-readme.md b/Plugin/unraid/pages/readme/monitor-readme.md
new file mode 100644
index 0000000..ceab52f
--- /dev/null
+++ b/Plugin/unraid/pages/readme/monitor-readme.md
@@ -0,0 +1,115 @@
+# Monitor
+
+Everything this host is doing right now, plus each partner. Nothing on this tab changes the
+system except the container controls — the rest is observation.
+
+Two poll rates, deliberately split: CPU, memory and network refresh every second; everything
+else every five. The slower payload is served from a cache the background writer refreshes once
+a minute, so polling faster could not make it newer — it only decides how soon the page notices.
+
+**Cards you do not have do not appear.** No GPU, no UPS, no VMs, no partner — the card is absent
+rather than showing zeros. An empty card would be a permanent reminder of nothing.
+
+---
+
+## Reference — System, power, CPU, memory, network
+
+| Card | What it tells you |
+|---|---|
+| System | Hostname, Unraid version, uptime, array state |
+| Power | UPS load, battery and runtime. Absent when no UPS is configured |
+| CPU | Per-core utilisation with a 2-minute history, plus package temperature |
+| Memory | Used / cached / free, broken out rather than a single percentage |
+| Network | Live rx/tx with a 2-minute history |
+
+## Reference — Jobs, fallback, partner, containers
+
+| Card | What it tells you |
+|---|---|
+| Scripts | What ran recently and how it exited. Click a row to open its log |
+| Fallback | `NORMAL`, `FALLBACK`, `NO_INTERNET` or `DARK` — which tier this host is in |
+| Partner | Each partner's uptime, load, memory and array state, over the mesh |
+| Containers & VMs | Grouped by folder. Start and stop from here; stopping asks first |
+
+## Reference — Sync, GPUs, transcodes, streams
+
+| Card | What it tells you |
+|---|---|
+| Rsync | Last run per share, duration and outcome |
+| GPU | Utilisation, VRAM and temperature, one card per GPU |
+| Transcode | Ramdisk usage, SSD fallback, and the sessions currently running |
+| Streams | What is playing now across Emby, Jellyfin and Plex |
+
+## Reference — Health and storage
+
+| Card | What it tells you |
+|---|---|
+| Watchdog | Roll-up across every watchdog. Healthy requires *every* strike set to be clear |
+| Parity | Last check, its result, and when the next one is due |
+| Pools | Each cache pool with its per-disk breakdown and live I/O |
+| Array | Every array disk, its temperature against the warning thresholds, and live I/O |
+
+## Reference — AI
+
+Present only on the AI host with `$AI_ENABLED` true. Absent everywhere else, on the same rule
+as the GPU and UPS cards.
+
+| Card | What it tells you |
+|---|---|
+| AI | Model residency, context size, VRAM, index size and staleness, tokens used today |
+| Conversations | The last `$AI_CHAT_HISTORY_MAX` chats. Click to reopen, × to delete |
+| Assistant | Ask a question without leaving the dashboard |
+
+---
+
+## The AI card leads with offload, not with size
+
+`GPU offload` is the number worth watching. At 100% the model is fully resident and answers at
+roughly 60 tokens per second on this hardware; with even a few layers pushed to CPU that falls
+to about a quarter of it. Nothing else in the WebGUI surfaces it, and a model that has quietly
+fallen back to partial offload is invisible until answers merely *feel* slow.
+
+`Index` reports chunks and whether the index is stale. Stale means a tracked file is newer than
+the last build — the assistant will still answer confidently, out of code that has since changed.
+That is the one failure a grounded answer cannot reveal on its own, so it is stated rather than
+implied.
+
+These figures come from the same one-minute collection the AI tab's banner uses, so the two
+cannot disagree about whether the model is loaded.
+
+## The assistant here starts on General Chat
+
+The AI tab starts on Varaverk Assistant, which answers only from this installation's
+documentation and cites its sources. This card starts on General Chat instead, because the jobs
+are different: the tab is where you go to interrogate the system, this is the box you type an
+idle question into while watching the dashboard.
+
+Nothing is lost by starting loose. A question that is genuinely about this installation is handed
+to the strict profile automatically, and handed back if the documentation turns out not to cover
+it. Starting strict would refuse ordinary questions to guard against a mistake the server already
+prevents.
+
+Switching profile with the buttons keeps what is on screen but stops sending earlier turns to the
+model. Carrying cited, retrieval-grounded answers into a mode with no retrieval makes the model
+keep referring to sources it can no longer see.
+
+## Conversations are shared with the AI tab
+
+The same store backs both. Start a thread here, finish it on the AI tab, or the reverse — it is
+one conversation, not two. Chats save themselves when a turn completes; there is no Save button,
+because a conversation worth keeping is not reliably one you knew was worth keeping while you
+were having it. The oldest drops off past `$AI_CHAT_HISTORY_MAX`.
+
+Reopening a chat shows the turns as plain text. Sources, reasoning and timings describe one
+generation and are not stored — redrawing them beside a transcript you may continue under a
+different profile would be citing evidence for an answer that is no longer being made.
+
+## What this tab will not do
+
+It will not clear a watchdog strike, reset a counter, or start a parity check. Strike state
+belongs to the watchdog that owns it, and a page that cleared strikes would silently undo an
+escalation something was deliberately building toward.
+
+Stopping a container asks first. Starting one does not — and either way the endpoint validates
+the request against real inventory, so a stale page cannot act on a container that no longer
+exists.