diff --git a/src/codex/subagent-model-fallback.ts b/src/codex/subagent-model-fallback.ts
index 27eea1e53d..ddd5b67e60 100644
--- a/src/codex/subagent-model-fallback.ts
+++ b/src/codex/subagent-model-fallback.ts
@@ -7,7 +7,7 @@
  */
 import { existsSync, readdirSync, readFileSync } from "node:fs";
 import { join } from "node:path";
-import { hasOwnProvider } from "../config";
+import { DEFAULT_SUBAGENT_MODELS, hasOwnProvider } from "../config";
 import { isRateLimitOrQuotaFailureMessage } from "../lib/errors";
 import type { OcxParsedRequest, OcxConfig } from "../types";
 import { slugsEquivalent } from "../providers/slug-codec";
@@ -609,9 +609,14 @@ export function applySubagentModelFallback(
   resolvedFallbackChain?: readonly string[] | null,
 ): { from?: string; to?: string; skipped?: string[] } | null {
   if (!isThreadSpawnRequest(headers)) return null;
-  const fallbackChain = resolvedFallbackChain === undefined
+  const configuredFallbackChain = resolvedFallbackChain === undefined
     ? resolveSubagentFallbackChain(parsed, config)
     : resolvedFallbackChain;
+  // Native-only encrypted V2 tasks need a readable ChatGPT backend even when the
+  // operator configured no fallback chain. Keep ordinary routed spawns unchanged.
+  const fallbackChain = configuredFallbackChain === null && nativeFallbackOnly
+    ? normalizedChain(parsed.modelId, config, [], DEFAULT_SUBAGENT_MODELS)
+    : configuredFallbackChain;
   if (!fallbackChain) return null;
   const selection = selectAvailableSubagentModel(
     parsed.modelId,
diff --git a/tests/subagent-model-fallback.test.ts b/tests/subagent-model-fallback.test.ts
index 75b2359ad5..d600072798 100644
--- a/tests/subagent-model-fallback.test.ts
+++ b/tests/subagent-model-fallback.test.ts
@@ -1056,6 +1056,29 @@ describe("subagent model fallback chain", () => {
     expect((parsed._rawBody as { model?: string }).model).toBe("alibaba-token-plan/qwen3.8-max");
   });
 
+  test("encrypted routed spawn gets an automatic native fallback when none is configured", () => {
+    const config = cfg({
+      subagentModelFallback: undefined,
+      defaultProvider: "xai",
+    });
+    const parsed = {
+      modelId: "xai/grok-4.5",
+      options: {},
+      context: { messages: [] },
+      _rawBody: { model: "xai/grok-4.5" },
+    };
+    const result = applySubagentModelFallback(
+      parsed as never,
+      new Headers({ "x-openai-subagent": "collab_spawn" }),
+      config,
+      "pool-a",
+      Date.now(),
+      true,
+    );
+    expect(result?.to).toBe("gpt-5.5");
+    expect(parsed.modelId).toBe("gpt-5.5");
+  });
+
   test("applySubagentModelFallback is a no-op for main turns", () => {
     updateAccountQuota("pool-a", 95);
     const parsed = {