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 v2.1.232 Home All releases ←olderv2.1.231 v2.1.233newer→
Claude Code v2.1.232

Subagent output scrubber now catches look-alike Unicode in fake control tags

You'll notice
Useful3 Signal2
Prompt Injection

Subagent output filtering now sees through homoglyphs and invisible characters in fake tags.

What

The filter that neutralizes instruction-shaped text coming back from a subagent was rebuilt to see through disguises: homoglyph letters, alternate dash and colon characters, and invisible formatting characters inserted between letters no longer get a fake tag past it. It runs on every subagent result with no flag around it.

Details
  • Covers the same five rules as before: system reminder tags, harness envelope tags, channel source tags, marker prefix forgery, and model layer tags. Previously each was a hand-written ASCII regex plus literal constants for antml: and <channel source=".
  • The neutralizer now appends a backslash to whatever matched, rather than rewriting a literal <.
  • A guard in the pattern builder rejects tag names outside lowercase [a-z0-9_-].
  • The warning banner shown to you is unchanged and still describes the substitution as < becoming <\.
Evidence

confusableTagScrubPattern: tag names are lowercase [a-z0-9_-]

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

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

See this entry in the whole of v2.1.232 →