A new settings panel that analyzes your local sessions and shows what's contributing to your rate limit usage, with actionable recommendations.
- Available to Pro and Max subscribers
- Shows approximate usage breakdown based on local sessions (does not include other devices or claude.ai)
- Tracks five cost behavior categories:
- Cache misses: "X% of your usage hit a >100k-token cache miss" — recommendation:
/compactbefore stepping away - Long context: "X% of your usage was at >150k context" — recommendation:
/compactmid-task,/clearwhen switching - Subagent-heavy sessions: "X% of your usage came from subagent-heavy sessions" — recommendation: configure cheaper models for simpler subagents
- High parallelism: "X% of your usage was while 4+ sessions ran in parallel" — recommendation: queue sessions for more even usage
- Long-running sessions: "X% of your usage came from sessions active for 8+ hours" — recommendation: ensure continuous usage is intentional
- Toggle between 24-hour and 7-day views with
dandwkeyboard shortcuts - Notes: "these are independent characteristics of your usage, not a breakdown"
- Warns about oversized session files (>200MB) that would skew analysis
Feature gated by tengu_birch_compass flag (search for "tengu_birch_compass" and "What's contributing to your limits usage")
Strings lifted out of the shipped bundle, so the claim above can be checked against them.
tengu_birch_compass On for this account, and not off by defaultThe flag server returned on for the one account this site reads, and nothing in this release compiles it off by default. The compiled default is shown below, and says which it is when we cannot read one: a fifth of gates compile in a string or a number rather than on or off, and most published releases have no gate table behind them at all. No client can see what the server returns for your account.
This account: on · anonymous baseline: off · compiled default in v2.1.108: 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 v2.1.108. 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.