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

One registry now backs every "warn about this once" case

Under the hood
Useful1 Signal1
Diagnostics

All warn-once diagnostics are now suppressed the same way through one registry.

What

A shared claim-once registry replaces the assorted booleans and sets that each warn-once site kept for itself, so repeated diagnostics are suppressed the same way everywhere.

Details
  • Callers now include the MCP tree-id tripwire under the key mcp_tree_id_tripwire, bare-primitive tool-result warnings keyed per tool, tool search, ink tree depth, frontmatter shadowing, the advisor unranked-base-model warning and the gateway retention sweep permission warning.
  • One dialog class dropped its own claim and reset pair in favour of a simple marker method.
Evidence

mcp_tree_id_tripwire, advisor_unranked_base_warning

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 →