Artifact tool results now flag co-writer or type-published content as untrusted data.
The asset-list schema (ge) gains a cowritten boolean field, and a new schema (Te) validates { assets: [...].extend({ from_id: ... }) } capped at a maximum count, laying groundwork for attributing assets copied from another artifact. The artifact file_list and file_read tool responses can now include a from_type flag when the served content type was determined by the artifact's declared type rather than sniffed.
Tool-result messages for asset_list, asset_read, and file_read now surface the cowritten field (and, for file listings, from_type), and the returned text explicitly warns the model to treat such assets or files as untrusted data when a co-writer or an Artifact type's publisher, rather than the user, produced them.