Source Intelligence

DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.

All of v2.1.246 Home All releases olderv2.1.245 v2.1.247newer
Claude Code v2.1.246

Artifact fetch responses are screened before the auth-error branch

Under the hood
Useful2 Signal2
Artifacts

Artifact downloads now pass through a screening step that can stop them with a canned message.

What

Both artifact fetch paths now run the raw status, headers and body through a shared screening step that can stop the fetch early with a canned message. It runs only for direct fetches and is skipped when the request goes through the cloud relay. The screening code and its map from reason to message live in a shared chunk, so which responses it rejects is not visible from the fetch paths themselves.

Evidence

[artifact] asset fetch

Strings lifted out of the shipped bundle, so the claim above can be checked against them.

See this entry in the whole of v2.1.246 →