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 v2.0.73 Home All releases olderv2.0.72 v2.0.74newer
Claude Code v2.0.73

Prompt Suggestion Filtering Enhanced

Significantly improved filtering of unsuitable prompt suggestions with new rejection criteria:

  • Word count limits: Suggestions must be 2-8 words
  • No evaluative phrases: Filters out "thanks", "looks good", "sounds good", etc.
  • No Claude voice: Rejects suggestions starting with "let me", "I'll", "here's", etc.
  • No multiple sentences: Suggestions with sentence boundaries are rejected
  • Added early exit conditions for pending permissions, active elicitations, and plan mode
Evidence

km5() at line 404234 (prompt suggestion filter, contains "claude_voice", "evaluative", "too_few_words")

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

See this entry in the whole of v2.0.73 →