Give the AI subsystem one profile table and one collection, read everywhere

Plugin/unraid/api/ai.php

@@ -102,14 +102,9 @@ header('Content-Type: application/json');
 header('Cache-Control: no-store, no-cache');
 require_once dirname(__DIR__) . '/include/ai.php';
 
-// History depth is per profile, and decided here rather than by the page. Varaverk Assistant
-// spends ~2500 of its 16384 on retrieved passages, so it cannot afford deep history; the other
-// two retrieve nothing and can carry a real conversation. Reasoning is not stored in history,
-// so it does not compound.
-// troubleshoot carries a whole log tail into context, so its history is the shallowest of the
-// four — the evidence for "why did this fail" is the log in front of it, not the conversation.
-const VV_AI_PROFILES = ['varaverk' => 3, 'chat' => 8, 'code' => 4, 'troubleshoot' => 2];
-const VV_AI_MAX_TURNS    = 3;     // fallback when a profile is not recognised
+// History depth is per profile and still decided server-side rather than by the page — the page
+// simply no longer carries a second copy of the numbers. include/ai_profiles.php holds them.
+// Reasoning is not stored in history, so it does not compound.
 const VV_AI_MAX_QUESTION = 4000;  // characters
 const VV_AI_MAX_HIST_MSG = 4000;  // characters per retained message
 const VV_AI_JOB_TTL      = 3600;  // seconds before a job file is reaped
@@ -153,8 +148,16 @@ if (!vv_ai_enabled()) {
 }
 
 // ── stats ─────────────────────────────────────────────────────────────────────
+// Served from the shared 'ai' cache that Tools/api_cache_writer.sh refreshes every minute, on
+// the same terms as the monitor and arrs payloads. This action is polled every 30 seconds by
+// every open tab and used to pay a full collection each time — around a second, most of it spent
+// waiting on Ollama and nvidia-smi — for numbers that only change when the writer next runs.
+//
+// ?live=1 bypasses it, for the case where something was just changed and the point is to see the
+// result. A missing cache always falls back to collecting, so the cache can never be the reason
+// the banner fails to render.
 if ($action === 'stats') {
-    echo json_encode(['ok' => true, 'stats' => vv_ai_stats()]);
+    echo json_encode(['ok' => true, 'stats' => vv_ai_stats_cached(isset($_GET['live']))]);
     exit;
 }
 
@@ -232,7 +235,7 @@ if ($action === 'chat_save') {
     if (!$isPost) { http_response_code(405); echo json_encode(['ok' => false, 'error' => 'POST only']); exit; }
 
     $profile = trim($_POST['profile'] ?? 'chat');
-    if (!isset(VV_AI_PROFILES[$profile])) {
+    if (!vv_ai_profile_ok($profile)) {
         echo json_encode(['ok' => false, 'error' => 'Unknown profile: ' . $profile]); exit;
     }
 
@@ -250,7 +253,7 @@ if ($action === 'chat_save') {
     // saved under one profile can be reopened under another, and the reopened turn is trimmed
     // again on the way back out by ask — so storing a little more than any single profile will
     // send costs nothing and keeps the transcript readable.
-    $cap = max(VV_AI_PROFILES) * 2;
+    $cap = vv_ai_profiles_max_turns() * 2;
     if (count($clean) > $cap) $clean = array_slice($clean, -$cap);
 
     $r = vv_ai_chat_save(trim($_POST['id'] ?? ''), $profile, $clean);
@@ -308,10 +311,10 @@ if ($action === 'ask') {
     }
 
     $profile = trim($_POST['profile'] ?? 'varaverk');
-    if (!isset(VV_AI_PROFILES[$profile])) {
+    if (!vv_ai_profile_ok($profile)) {
         echo json_encode(['ok' => false, 'error' => 'Unknown profile: ' . $profile]); exit;
     }
-    $maxTurns = VV_AI_PROFILES[$profile] ?? VV_AI_MAX_TURNS;
+    $maxTurns = vv_ai_profile_turns($profile);
 
     // Where the caller is standing — "master.conf", "daily_sync_maintenance.sh", a log name.
     // The scheduler page sends it so a question can say "this setting" and mean something; the

Plugin/unraid/api/monitor.php

@@ -100,15 +100,14 @@ vv_api_data();
 //
 // Null on any host that is not the AI host or has AI_ENABLED false, which is also what makes the
 // row absent rather than empty there. Same shape as every other optional subsystem on this page.
+// Reads the shared 'ai' cache the writer maintains and only collects on a miss. Landing here at
+// all already means the monitor cache missed; paying a second full AI collection on top of that
+// would make the slowest request on this page slower still, for figures a background writer
+// refreshed under a minute ago.
 $_vv_ai = null;
 if (vv_ai_ui_on()) {
     require_once dirname(__DIR__) . '/include/ai.php';
-    $_vv_ai = [
-        'model'   => vv_ai_config()['model'],
-        'runtime' => vv_ai_runtime_stats(),
-        'index'   => vv_ai_index_stats(),
-        'tokens'  => vv_ai_token_stats()['today'] ?? null,
-    ];
+    $_vv_ai = vv_ai_monitor_block(vv_ai_stats_cached(isset($_GET['live'])));
 }
 
 echo json_encode([