feat(combo): response quality validation, circuit breaker fix, Cursor 4.6 models (#762)

- Add `validateResponseQuality()` to detect empty/invalid 200 responses from
  upstream providers in combo routing. Non-streaming responses with empty body,
  invalid JSON, or missing content/tool_calls now trigger circuit breaker
  failure and fallback to the next model instead of being returned to the client.

- Add missing `breaker._onSuccess()` calls in both priority and round-robin
  combo paths. Previously failures accumulated without reset, causing premature
  circuit breaker trips on healthy models.

- Update Cursor provider registry with Claude 4.6 model IDs (opus-high,
  sonnet-high, haiku, opus + thinking variants). Keep 4.5 IDs for backward
  compatibility.

- Update free-stack preset: replace duplicate qw/qwen3-coder-plus with
  if/deepseek-v3.2 for better model diversity.

- Add paid-premium combo template for round-robin load distribution across
  paid subscription providers (Cursor, Antigravity).

Made-with: Cursor

open-sse/config/providerRegistry.ts

@@ -500,6 +500,12 @@ export const REGISTRY: Record<string, RegistryEntry> = {
     clientVersion: "1.1.3",
     models: [
       { id: "default", name: "Auto (Server Picks)" },
+      { id: "claude-4.6-opus-high-thinking", name: "Claude 4.6 Opus High Thinking" },
+      { id: "claude-4.6-opus-high", name: "Claude 4.6 Opus High" },
+      { id: "claude-4.6-sonnet-high-thinking", name: "Claude 4.6 Sonnet High Thinking" },
+      { id: "claude-4.6-sonnet-high", name: "Claude 4.6 Sonnet High" },
+      { id: "claude-4.6-haiku", name: "Claude 4.6 Haiku" },
+      { id: "claude-4.6-opus", name: "Claude 4.6 Opus" },
       { id: "claude-4.5-opus-high-thinking", name: "Claude 4.5 Opus High Thinking" },
       { id: "claude-4.5-opus-high", name: "Claude 4.5 Opus High" },
       { id: "claude-4.5-sonnet-thinking", name: "Claude 4.5 Sonnet Thinking" },

open-sse/services/combo.ts

@@ -45,6 +45,80 @@ const DEFAULT_MODEL_P95_MS = {
 };
 const MIN_HISTORY_SAMPLES = 10;
 
+/**
+ * Validate that a successful (HTTP 200) non-streaming response actually contains
+ * meaningful content. Returns { valid: true } or { valid: false, reason }.
+ *
+ * Only inspects non-streaming JSON responses — streaming responses are passed through
+ * because buffering the full stream would defeat the purpose of streaming.
+ *
+ * Checks:
+ * 1. Body is valid JSON
+ * 2. Has at least one choice with non-empty content or tool_calls
+ */
+async function validateResponseQuality(
+  response: Response,
+  isStreaming: boolean,
+  log: { warn?: (...args: any[]) => void }
+): Promise<{ valid: boolean; reason?: string; clonedResponse?: Response }> {
+  if (isStreaming) return { valid: true };
+
+  const contentType = response.headers.get("content-type") || "";
+  if (!contentType.includes("application/json") && !contentType.includes("text/")) {
+    return { valid: true };
+  }
+
+  let cloned: Response;
+  try {
+    cloned = response.clone();
+  } catch {
+    return { valid: true };
+  }
+
+  let text: string;
+  try {
+    text = await cloned.text();
+  } catch {
+    return { valid: true };
+  }
+
+  if (!text || text.trim().length === 0) {
+    return { valid: false, reason: "empty response body" };
+  }
+
+  let json: any;
+  try {
+    json = JSON.parse(text);
+  } catch {
+    if (text.startsWith("data:")) return { valid: true };
+    return { valid: false, reason: "response is not valid JSON" };
+  }
+
+  const choices = json?.choices;
+  if (!Array.isArray(choices) || choices.length === 0) {
+    if (json?.output || json?.result || json?.data || json?.response) return { valid: true };
+    if (json?.error) return { valid: false, reason: `upstream error in 200 body: ${json.error?.message || JSON.stringify(json.error).substring(0, 200)}` };
+    return { valid: true };
+  }
+
+  const firstChoice = choices[0];
+  const message = firstChoice?.message || firstChoice?.delta;
+  if (!message) {
+    return { valid: false, reason: "choice has no message object" };
+  }
+
+  const content = message.content;
+  const toolCalls = message.tool_calls;
+  const hasContent = content !== null && content !== undefined && content !== "";
+  const hasToolCalls = Array.isArray(toolCalls) && toolCalls.length > 0;
+
+  if (!hasContent && !hasToolCalls) {
+    return { valid: false, reason: "empty content and no tool_calls in response" };
+  }
+
+  return { valid: true };
+}
+
 // In-memory atomic counter per combo for round-robin distribution
 // Resets on server restart (by design — no stale state)
 const rrCounters = new Map();
@@ -872,14 +946,31 @@ export async function handleComboChat({
 
       const result = await handleSingleModelWrapped(body, modelStr);
 
-      // Success — return response
+      // Success — validate response quality before returning
       if (result.ok) {
+        const quality = await validateResponseQuality(result, !!body.stream, log);
+        if (!quality.valid) {
+          log.warn(
+            "COMBO",
+            `Model ${modelStr} returned 200 but failed quality check: ${quality.reason}`
+          );
+          breaker._onFailure();
+          recordComboRequest(combo.name, modelStr, {
+            success: false,
+            latencyMs: Date.now() - startTime,
+            fallbackCount,
+            strategy,
+          });
+          if (i > 0) fallbackCount++;
+          break; // move to next model
+        }
         resolvedByModel = modelStr;
         const latencyMs = Date.now() - startTime;
         log.info(
           "COMBO",
           `Model ${modelStr} succeeded (${latencyMs}ms, ${fallbackCount} fallbacks)`
         );
+        breaker._onSuccess();
         recordComboRequest(combo.name, modelStr, {
           success: true,
           latencyMs,
@@ -1139,13 +1230,30 @@ async function handleRoundRobinCombo({
 
         const result = await handleSingleModel(body, modelStr);
 
-        // Success
+        // Success — validate response quality before returning
         if (result.ok) {
+          const quality = await validateResponseQuality(result, !!body.stream, log);
+          if (!quality.valid) {
+            log.warn(
+              "COMBO-RR",
+              `${modelStr} returned 200 but failed quality check: ${quality.reason}`
+            );
+            breaker._onFailure();
+            recordComboRequest(combo.name, modelStr, {
+              success: false,
+              latencyMs: Date.now() - startTime,
+              fallbackCount,
+              strategy: "round-robin",
+            });
+            if (offset > 0) fallbackCount++;
+            break; // move to next model
+          }
           const latencyMs = Date.now() - startTime;
           log.info(
             "COMBO-RR",
             `${modelStr} succeeded (${latencyMs}ms, ${fallbackCount} fallbacks)`
           );
+          breaker._onSuccess();
           recordComboRequest(combo.name, modelStr, {
             success: true,
             latencyMs,

src/app/(dashboard)/dashboard/combos/page.tsx

@@ -186,6 +186,9 @@ const COMBO_TEMPLATE_FALLBACK = {
   freeStackTitle: "Free Stack ($0)",
   freeStackDesc:
     "Round-robin across all free providers: Kiro, iFlow, Qwen, Gemini CLI. Zero cost, never stops.",
+  paidPremiumTitle: "Paid Premium",
+  paidPremiumDesc:
+    "Round-robin across paid subscriptions: Cursor, Antigravity. Top-tier models, distributed load.",
 };
 
 const COMBO_TEMPLATES = [
@@ -250,6 +253,21 @@ const COMBO_TEMPLATES = [
       healthCheckEnabled: true,
     },
   },
+  {
+    id: "paid-premium",
+    icon: "workspace_premium",
+    titleKey: "templatePaidPremium",
+    descKey: "templatePaidPremiumDesc",
+    fallbackTitle: COMBO_TEMPLATE_FALLBACK.paidPremiumTitle,
+    fallbackDesc: COMBO_TEMPLATE_FALLBACK.paidPremiumDesc,
+    strategy: "round-robin",
+    suggestedName: "paid-premium",
+    config: {
+      maxRetries: 2,
+      retryDelayMs: 1000,
+      healthCheckEnabled: true,
+    },
+  },
 ];
 
 function getStrategyMeta(strategy) {
@@ -1425,18 +1443,27 @@ function ComboFormModal({ isOpen, combo, onClose, onSave, activeProviders }) {
     { model: "kr/claude-sonnet-4.5", weight: 0 },
     { model: "if/kimi-k2-thinking", weight: 0 },
     { model: "if/qwen3-coder-plus", weight: 0 },
-    { model: "qw/qwen3-coder-plus", weight: 0 },
+    { model: "if/deepseek-v3.2", weight: 0 },
     { model: "nvidia/llama-3.3-70b-instruct", weight: 0 },
     { model: "groq/llama-3.3-70b-versatile", weight: 0 },
   ];
 
+  const PAID_PREMIUM_PRESET_MODELS = [
+    { model: "cu/claude-4.6-opus-high", weight: 0 },
+    { model: "ag/claude-sonnet-4-6", weight: 0 },
+    { model: "cu/claude-4.6-sonnet-high", weight: 0 },
+    { model: "ag/gpt-5", weight: 0 },
+    { model: "ag/gemini-3.1-pro-preview", weight: 0 },
+  ];
+
   const applyTemplate = (template) => {
     setStrategy(template.strategy);
     setConfig((prev) => ({ ...prev, ...template.config }));
     if (!name.trim()) setName(template.suggestedName);
-    // Pre-fill Free Stack with 7 real free provider models
     if (template.id === "free-stack") {
       setModels(FREE_STACK_PRESET_MODELS);
+    } else if (template.id === "paid-premium") {
+      setModels(PAID_PREMIUM_PRESET_MODELS);
     }
   };