Version Base Tracking:
- New
version_basetelemetry field normalizes version strings (e.g., "2.0.37-beta" → "2.0.37") - Enables cleaner analytics aggregation across pre-release versions
- Evidence:
Jo5()function at line 449429,versionBasemapping at line 449396
Improved Batch Configuration:
- Telemetry export interval increased from 5 seconds to 20 seconds (4x longer)
- New batch size limit of 200 events per export
- Remote configuration support via
tengu_1p_event_batch_configdynamic config - Reduces network overhead and improves performance
- Evidence:
xBI = 20000andvBI = 200at lines 478225-478226, dynamic config iner2()at lines 478169-478173
Rejection Feedback Tracking:
- New
hasFeedbackfield tracks whether users provided feedback when rejecting tools - Enables analysis of feature adoption and user engagement
- Evidence:
QZ()function at line 352406,hasFeedbackparameter at lines 1000-1002
Feature flag
tengu_1p_event_batch_config Not enough to sayNothing here resolved what this flag was doing on this version, so nothing here should be read as on or off.
This account: no value returned · anonymous baseline: no value returned · compiled default in v2.0.37: no gate table built for this version
These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.
Read once, for one account on one subscription tier, against v2.0.37. It isn't a statement about your account. What a flag value here can and cannot tell you
Flag reading moved
The flag server now returns {"maxExportBatchSize":400,"maxQueueSize":8192,"path":"/api/event_logging/v2/batch","scheduledDelayMillis":30000} for tengu_1p_event_batch_config, read as this account.
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.
Flag reading moved
The flag server now returns nothing at all, having stopped carrying it for tengu_1p_event_batch_config, read as this account.
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.