Follow Discord
Sweep 22 Oct 2026 · 04:32Z Build v2.1.308 527 read Stable v2.1.308 Latest v2.1.308 Next v2.1.308 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.307 ·

Claude's subagent instructions now stress what gets lost in a handoff

In the agent-claude prompt, the Agent tool's description now warns that detail is lost both when Claude briefs a subagent and when it reports back

You'll notice Improvements
JSON All of v2.1.307
You'll noticeTier: how much it should matter to you
2Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
SubagentsArea: what it touches
ImprovementsKind: in v2.1.307,
ImprovementsSection of the release
What

The Agent tool lets Claude hand a task to a subagent, a separate helper that works on its own and returns a summary. Its description in the agent-claude prompt has been rewritten.

The old text mostly told Claude when to delegate: matching agent types, parallel work, reading across many files. The new text adds a warning about the two handoffs. It says the subagent knows only what Claude puts in the prompt, Claude sees only the summary that comes back, and detail is dropped both ways without either side knowing what the other missed.

The standard Agent tool description is unchanged.

Why

These are instructions Claude follows, so they can change how often it delegates and how fully it briefs a subagent. That effect has not been measured.

See this entry in the whole of v2.1.307 →

Feedback