Source Intelligence
Follow Discord

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 v2.1.225 Home All releases ←olderv2.1.224 v2.1.226newer→
Claude Code v2.1.225

Classifier refusals are tagged separately in blocking telemetry

Under the hood
Useful1 Signal0
Telemetry

Safety-classifier refusals are now logged distinctly from ordinary parse failures.

What

The two-stage prompt-injection and safeguard classifier now records when the model stopped with a refusal, which previously looked identical to an ordinary parse failure.

Details
  • The parse-failure event now carries stopReason plus failureKind: "policy_refusal".
  • The block result itself carries refusedBySafeguard.
Evidence

refusedBySafeguard

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

See this entry in the whole of v2.1.225 →