An intelligent context window management system that automatically removes older tool results to stay within context limits, triggered by server-side hints.
Feature-flagged via tengu_hazel_osprey (default: false), with beta header context-hint-2026-04-09
- When the context window approaches limits, old tool result content is replaced with
"[Old tool result content cleared]" - Handles HTTP 422/424 status codes as context-hint rejection signals
- Tracks telemetry for
tengu_time_based_microcompactwith token savings metrics - Clears cached thinking tokens on context-hint triggers
Microcompact system (gated by tengu_hazel_osprey, search for "[Old tool result content cleared]" and "tengu_time_based_microcompact")
Strings lifted out of the shipped bundle, so the claim above can be checked against them.
tengu_hazel_osprey Off in both readingsThe flag server returned off for the account this site reads and for the anonymous baseline. A reading of off cannot rule out a rollout these two readings sit outside of.
This account: off · anonymous baseline: off · compiled default in v2.1.110: 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 v2.1.110. It isn't a statement about your account. What a flag value here can and cannot tell you
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.
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.