What
- Artifact reads now track whether an artifact is
external, meaning it was shared from a different organization than your own, and show distinct messaging for that case ("shared with you from another organization" / noting the author is outside your organization), separate from the existing public/owner/writer distinctions. - When artifact data comes back marked with mode
external, an explicitexternal: trueflag is now passed along through both the artifact-read and comments-list code paths, rather than being dropped. - A new
typePageparameter (defaulting to"none") controls whether the artifact's "type contract" text (a description of the artifact's structure) is fetched and included when reading it, so that text isn't refetched on every single read.
Why
This makes it clearer to users and to Claude when an artifact originated outside their own organization, and avoids unnecessary repeated fetching of type-contract text on every artifact read.
Feature flag
tengu_cobalt_plinth_larch 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: off · compiled default in v2.1.265: off
Read once, for one account on one subscription tier, against v2.1.265. It isn't a statement about your account. What a flag value here can and cannot tell you
Flag reading moved
The flag server now returns on for tengu_cobalt_plinth_larch, 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 off for tengu_cobalt_plinth_larch, 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 on for tengu_cobalt_plinth_larch, 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.