Follow Discord
Sweep 08 Oct 2026 · 18:53Z Build v2.1.295 516 read Stable v2.1.286 Latest v2.1.295 Next v2.1.295 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.295 ·

Screened prompts are saved to history as they were actually sent

When a prompt is screened, history and the transcript now record the rewritten text, or nothing if the prompt was dropped

You'll notice Improvements
JSON All of v2.1.295
You'll noticeTier: how much it should matter to you
1Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
Prompt InputArea: what it touches
ImprovementsKind: in v2.1.295,
ImprovementsSection of the release

Unclear What causes a prompt to be screened, and whether screening runs for a given user, is not known.

What

Some prompts are screened (checked before they go to the model). Claude Code now decides what to save from the screening result. This applies both to your input history (the list of earlier prompts you can recall) and to the session transcript.

  • A dropped prompt: no history entry is written.
  • A rewritten prompt: the rewritten text is stored, not the text you typed.
Why

The text saved in history and in the transcript can now differ from what you typed. Recalling an earlier prompt may bring back the rewritten version, or nothing at all.

How sure we are
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubtWhat causes a prompt to be screened, and whether screening runs for a given user, is not known.

See this entry in the whole of v2.1.295 →

Feedback