Follow Discord
Sweep 22 Oct 2026 · 04:32Z Build v2.1.308 527 read Stable v2.1.308 Latest v2.1.308 Next v2.1.308 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.75 ·

File Read Size Controlled by Feature Flag

Entry
JSON All of v2.1.75
EntryKind: in v2.1.75,
ChangesSection of the release

File read limits (maxSizeBytes and maxTokens) are now configurable via the tengu_amber_wren feature flag, allowing server-side tuning of read limits without client updates. A new CLAUDE_CODE_AUTO_COMPACT_WINDOW environment variable also allows users to override the auto-compaction window size.

Evidence

Feature-flag-controlled read limits (search for "tengu_amber_wren", "CLAUDE_CODE_AUTO_COMPACT_WINDOW")

Strings lifted out of the shipped bundle, so the claim above can be checked against them.

Read from
Feature flag
tengu_amber_wren Not enough to say

Nothing 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 v2.1.75: 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 v2.1.75. It isn't a statement about your account. What a flag value here can and cannot tell you

What has happened since
Flag reading moved The flag server now returns {"maxTokens":25000,"targetedRangeNudge":true} for tengu_amber_wren, 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_amber_wren, 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.

See this across every release →

See this entry in the whole of v2.1.75 →

Feedback