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 v2.1.238 Home All releases olderv2.1.237 v2.1.239newer
Claude Code v2.1.238

Artifact auto-reply messages distinguish a pause from a stop

You'll notice
Useful2 Signal1
Artifacts

Artifact auto-reply messages now describe interruptions as a pause that lifts on your next message.

What

Interrupting a watch is now described as a pause that lifts by itself when you send a message, and the refusal when there is nothing to resume says so directly. The hourly rate-limit message now says auto-reply is "held back" rather than "paused".

Details
  • The notice shown when watches are not resumed on session resume is now just the list of artifacts, without the longer explanation about resumed sessions carrying only the most recent monitor.
  • The still-arming case no longer reuses the stopped message.
  • Internally, a resumed auto-react ledger is restored alongside the comment monitor, and the auto-react scanner no longer bails early on an aborted signal, so scan errors are always reported.
Evidence

No auto-reply stop or pause is recorded for this artifact in this session, so there is nothing to resume (an interrupt's pause lifts on its own when the user sends a message)., auto-reply held back (hourly cap); use the Artifact tool to read and reply.

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

See this entry in the whole of v2.1.238 →