Follow Discord
Sweep 10 Nov 2026 · 02:18Z Build v2.1.322 539 read Stable v2.1.311 Latest v2.1.322 Next v2.1.322 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.281 ·

Tool-result size telemetry: token estimate, media count, will-persist flag

Usage data now records a tool result's estimated tokens, image and document count, and whether it will be saved to disk

Under the hood Internal Changes
JSON All of v2.1.281
Under the hoodTier: how much it should matter to you
1Useful: my rating, 1 to 5
1Signal: worth watching, 1 to 5
TelemetryArea: what it touches
Internal ChangesKind: in v2.1.281,
Internal ChangesSection of the release
What

A tool result is what a tool hands back to Claude after it runs. A new helper records three measurements of it in usage data:

  • toolResultTokensEst, an estimate of its size in tokens (the units the model reads text in)
  • toolResultMediaBlocks, how many images or documents it contains
  • toolResultWillPersist, whether it is large enough to be saved rather than kept inline

If the measurement fails, the error is ignored.

Why

This adds measurement only. A failure inside it never interrupts the tool call.

See this entry in the whole of v2.1.281 →

Feedback