Follow Discord
Sweep 22 Oct 2026 · 21:04Z Build v2.1.309 528 read Stable v2.1.308 Latest v2.1.309 Next v2.1.309 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.68 ·

Effort selection dialog redesigned

Entry
JSON All of v2.1.68
EntryKind: in v2.1.68,
ChangesSection of the release

The effort callout dialog for Opus 4.6 users has been redesigned with server-configurable content and a simplified option set.

Details
  • Dialog title and description are now fetched from the server (via tengu_grey_step2 feature flag) instead of being hardcoded
  • Options are simplified to three clear choices:
  • "Use medium effort (recommended)"
  • "Use high effort"
  • "Use low effort" (new — previously only medium and high were offered)
  • Removed verbose per-option descriptions and the "(current)" suffix labeling
  • Uses a separate dismiss state (effortCalloutV2Dismissed) so users who dismissed the old dialog may see the new one
Evidence

Server-controlled effort dialog (search for "tengu_grey_step2") — zb6() at line ~162660, dialog at ohq() line ~588325

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

Read from
Feature flag
tengu_grey_step2 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 v2.1.68: 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 v2.1.68. 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 {"dialogDescription":"Effort determines how long Claude thinks for when completing your task. We recommend medium effort for most tasks to balance speed and intelligence and maximize rate limits. Use ultrathink to trigger high effort when needed.","dialogTitle":"We recommend medium effort for Opus","enabled":true} for tengu_grey_step2, 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_grey_step2, 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 v2.1.68 →

Feedback