Follow Discord
Sweep 27 Oct 2026 · 19:04Z Build v2.1.312 531 read Stable v2.1.308 Latest v2.1.312 Next v2.1.312 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.312 ·

The note on partial output from a cut-off agent is now built a different way

Claude Code now builds the note marking a cut-off agent's recovered output in one shared place and passes on a filtered history

Under the hood Internal Changes
JSON All of v2.1.312
Under the hoodTier: how much it should matter to you
1Useful: my rating, 1 to 5
0Signal: worth watching, 1 to 5
SubagentsArea: what it touches
Internal ChangesKind: in v2.1.312,
Internal ChangesSection of the release
What

A subagent is a helper Claude starts to do part of a job. When one is cut off before finishing, Claude Code recovers what it had produced and adds a note saying everything below is partial output. That note is now built by a shared routine instead of being written out in place, and the conversation history passed on with it is the filtered one.

Why

The wording the model sees when a subagent is cut off may differ from before.

What the documentation says
e.message spans modified, high confidence | `gen_ai.input.messages` | `string` | <Badge>Content</Badge> A JSON string that follows the OpenTelemetry [`InputMessages`](https://github.com/open-telemetry/semantic-conventions-genai/blob/e57c543b4889619eb2a05702471937db5119165d/model/g… see the edit
How sure we are
One source agreesOne thing we can check says the same as this entry.
Anthropic's documentation agreese.message on spans

See this entry in the whole of v2.1.312 →

Feedback