Detect runtime retry signals from assistant text parts

src/hooks/runtime-fallback/index.test.ts

@@ -387,6 +387,92 @@ describe("runtime-fallback", () => {
       expect(fallbackLog?.data).toMatchObject({ from: "openai/gpt-5.3-codex", to: "anthropic/claude-opus-4-6" })
     })
 
+    test("should trigger fallback on auto-retry signal in assistant text parts", async () => {
+      const hook = createRuntimeFallbackHook(createMockPluginInput(), {
+        config: createMockConfig({ notify_on_fallback: false }),
+        pluginConfig: createMockPluginConfigWithCategoryFallback(["openai/gpt-5.2"]),
+      })
+
+      const sessionID = "test-session-parts-auto-retry"
+      SessionCategoryRegistry.register(sessionID, "test")
+
+      await hook.event({
+        event: {
+          type: "session.created",
+          properties: { info: { id: sessionID, model: "quotio/claude-opus-4-6" } },
+        },
+      })
+
+      await hook.event({
+        event: {
+          type: "message.updated",
+          properties: {
+            info: {
+              sessionID,
+              role: "assistant",
+              model: "quotio/claude-opus-4-6",
+            },
+            parts: [
+              {
+                type: "text",
+                text: "This request would exceed your account's rate limit. Please try again later. [retrying in 2s attempt #2]",
+              },
+            ],
+          },
+        },
+      })
+
+      const signalLog = logCalls.find((c) => c.msg.includes("Detected provider auto-retry signal"))
+      expect(signalLog).toBeDefined()
+
+      const fallbackLog = logCalls.find((c) => c.msg.includes("Preparing fallback"))
+      expect(fallbackLog).toBeDefined()
+      expect(fallbackLog?.data).toMatchObject({ from: "quotio/claude-opus-4-6", to: "openai/gpt-5.2" })
+    })
+
+    test("should trigger fallback when auto-retry text parts are nested under info.parts", async () => {
+      const hook = createRuntimeFallbackHook(createMockPluginInput(), {
+        config: createMockConfig({ notify_on_fallback: false }),
+        pluginConfig: createMockPluginConfigWithCategoryFallback(["openai/gpt-5.2"]),
+      })
+
+      const sessionID = "test-session-info-parts-auto-retry"
+      SessionCategoryRegistry.register(sessionID, "test")
+
+      await hook.event({
+        event: {
+          type: "session.created",
+          properties: { info: { id: sessionID, model: "quotio/claude-opus-4-6" } },
+        },
+      })
+
+      await hook.event({
+        event: {
+          type: "message.updated",
+          properties: {
+            info: {
+              sessionID,
+              role: "assistant",
+              model: "quotio/claude-opus-4-6",
+              parts: [
+                {
+                  type: "text",
+                  text: "This request would exceed your account's rate limit. Please try again later. [retrying in 2s attempt #2]",
+                },
+              ],
+            },
+          },
+        },
+      })
+
+      const signalLog = logCalls.find((c) => c.msg.includes("Detected provider auto-retry signal"))
+      expect(signalLog).toBeDefined()
+
+      const fallbackLog = logCalls.find((c) => c.msg.includes("Preparing fallback"))
+      expect(fallbackLog).toBeDefined()
+      expect(fallbackLog?.data).toMatchObject({ from: "quotio/claude-opus-4-6", to: "openai/gpt-5.2" })
+    })
+
     test("should trigger fallback on session.status auto-retry signal", async () => {
       const promptCalls: unknown[] = []
       const hook = createRuntimeFallbackHook(

src/hooks/runtime-fallback/message-update-handler.ts

@@ -57,10 +57,20 @@ export function createMessageUpdateHandler(deps: HookDeps, helpers: AutoRetryHel
   return async (props: Record<string, unknown> | undefined) => {
     const info = props?.info as Record<string, unknown> | undefined
     const sessionID = info?.sessionID as string | undefined
-    const retrySignalResult = extractAutoRetrySignal(info)
-    const retrySignal = retrySignalResult?.signal
     const timeoutEnabled = config.timeout_seconds > 0
-    const parts = props?.parts as Array<{ type?: string; text?: string }> | undefined
+    const eventParts = props?.parts as Array<{ type?: string; text?: string }> | undefined
+    const infoParts = info?.parts as Array<{ type?: string; text?: string }> | undefined
+    const parts = eventParts && eventParts.length > 0 ? eventParts : infoParts
+    const retrySignalResult = extractAutoRetrySignal(info)
+    const partsText = (parts ?? [])
+      .filter((p) => typeof p?.text === "string")
+      .map((p) => (p.text ?? "").trim())
+      .filter((text) => text.length > 0)
+      .join("\n")
+    const retrySignalFromParts = partsText
+      ? extractAutoRetrySignal({ message: partsText, status: partsText, summary: partsText })?.signal
+      : undefined
+    const retrySignal = retrySignalResult?.signal ?? retrySignalFromParts
     const errorContentResult = containsErrorContent(parts)
     const error = info?.error ?? 
       (retrySignal && timeoutEnabled ? { name: "ProviderRateLimitError", message: retrySignal } : undefined) ??