The /advisor command is now a full interactive dialog with rich descriptions, replacing the previous text-based interface.
/advisor # Opens the advisor selection dialog
/advisor opus # Set advisor directly
/advisor off # Disable advisor- New dialog title "Advisor Tool" with explanatory text about how the advisor works
- Explains: "When Claude needs stronger judgment — a complex decision, an ambiguous failure, a problem it's circling without progress — it escalates to the advisor model for guidance, then resumes"
- Highlights the Sonnet + Opus pairing: "near-Opus performance with reduced token usage"
- Shows a warning if the current main model doesn't support the advisor
- Valid advisor options are "opus" and "sonnet"
- Description changed from "Configure the advisor model" to "Configure the Advisor Tool to consult a stronger model for guidance at key moments during a task"
- Advisor enablement is no longer gated by
canUserConfigurefromtengu_sage_compass; now uses a simplernb()check
Advisor dialog redesign (search for "Configure the Advisor Tool to consult a stronger model for guidance at key moments during a task", "Advisor Tool", "When Claude needs stronger judgment")
Strings lifted out of the shipped bundle, so the claim above can be checked against them.
tengu_sage_compass 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 v2.1.92: 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 v2.1.92. 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.