You'll noticeTier: how much it should matter to you
1Useful: my rating, 1 to 5
1Signal: worth watching, 1 to 5
ToolsArea: what it touches
ImprovementsKind: in v2.1.293,
ImprovementsSection of the release
What
When Claude tries to use a tool it does not have, Claude Code sends back an error. That error used to say every time that the tool was disabled for the session, in subagents too. It now tells two cases apart:
The tool is disabled for this session.
The tool is not among the tools available here, and Claude should continue with the ones it has.
For the Agent tool, which starts subagents (helper copies of Claude that work on a side task), it also says "An agent cannot be continued from this context." and tells Claude to start a new call instead.
Why
Claude gets more accurate advice on how to recover after a bad tool call, so it is less likely to keep retrying something that cannot work.
How sure we are
One source agreesOne thing we can check says the same as this entry.
Anthropic's release notes agreeFixed a custom agent named worker being shown as "Agent" when it starts, and the agent detail dialog's title losing the agent type once the…
