Lower the default timeline window event budget to 400

apps/app/src/lib/system-config-atoms.ts

@@ -31,7 +31,7 @@ const unavailableSystemConfig: SystemConfigResponse = {
   appearance: defaultAppTheme,
   customThemes: [],
   pluginThemes: [],
-  featureFlags: { placeholder: false, timelineWindowEventBudget: 1_500 },
+  featureFlags: { placeholder: false, timelineWindowEventBudget: 400 },
   hostDaemonPort: null,
   localHelperPorts: [],
   serverUrl: "",

apps/server/test/services/threads/timeline-event-budget.test.ts

@@ -418,7 +418,7 @@ describe("timeline event budget", () => {
       page,
     };
     expect(
-      buildThreadTimeline(db, thread, { ...options, eventBudget: 1_500 }),
+      buildThreadTimeline(db, thread, { ...options, eventBudget: 400 }),
     ).toEqual(
       buildThreadTimeline(db, thread, {
         ...options,

docs/configuration.md

@@ -694,7 +694,7 @@ is synchronous, so it blocked the server's event loop — which also delayed
 dynamic tool call and before registering every interactive request. One slow
 thread therefore slowed agent work on _every_ thread on the host.
 
-A window is capped at `BB_FF_TIMELINE_WINDOW_EVENT_BUDGET` events (default 1500) and returns however many whole turns fit. Older turns load automatically
+A window is capped at `BB_FF_TIMELINE_WINDOW_EVENT_BUDGET` events (default 400) and returns however many whole turns fit. Older turns load automatically
 as you scroll toward the top of the loaded window; a manual "Load older
 messages" button remains on surfaces that render no scroll body, and after a
 failed page so a broken fetch is retried on request rather than in a loop.

packages/config/test/config.test.ts

@@ -310,7 +310,7 @@ describe("consumer-specific config", () => {
     expect(serverConfig.OPENAI_API_KEY).toBe("test-openai-key");
     expect(serverConfig.featureFlags).toEqual({
       placeholder: false,
-      timelineWindowEventBudget: 1_500,
+      timelineWindowEventBudget: 400,
     });
   });
 

packages/domain/src/feature-flags.ts

@@ -8,5 +8,5 @@ export type FeatureFlags = z.infer<typeof featureFlagsSchema>;
 
 export const defaultFeatureFlags: FeatureFlags = {
   placeholder: false,
-  timelineWindowEventBudget: 1_500,
+  timelineWindowEventBudget: 400,
 };

packages/templates/src/templates/bb-guide-customization.md

@@ -142,7 +142,7 @@ timelines and large expanded timeline details. Enable it with
 `bb settings experiment timelineWindowing true`.
 
 Thread timeline windows are bounded by event count as well as user-message
-count (`BB_FF_TIMELINE_WINDOW_EVENT_BUDGET`, default 1500), so a long thread
+count (`BB_FF_TIMELINE_WINDOW_EVENT_BUDGET`, default 400), so a long thread
 stops reprojecting its whole history — and blocking the server event loop — on
 every update. A turn still running is cut at the budget too, so a very long
 turn costs the budget per update instead of growing without limit. Older