Follow Discord
Sweep 12 Nov 2026 · 23:00Z Build v2.1.323 540 read Stable v2.1.311 Latest v2.1.323 Next v2.1.323 Feeds RSS JSON llms.txt llms-full.txt Unofficial
HomeReleasesv2.1.323entry #85

Claude Code v2.1.323 ·

Tool hooks receive more details about agents and MCP calls

Tool-use hook runs now get the agent's ID and type, and MCP tool calls carry more labels, including whether a call is not read-only

Use it now Improvements
JSON All of v2.1.323
Use it nowTier: how much it should matter to you
3Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
HooksArea: what it touches
ImprovementsKind: in v2.1.323,
What probably matters to youSection of the release

Unclear It is not stated whether these new details reach hooks you configure yourself.

What

Hooks are steps that run automatically around a tool call, which is any action Claude takes such as running a command. Claude Code now passes more detail along to them:

  • The ID and type of the agent that made the call.
  • For MCP tools, from outside services, a flag when the call is not read-only.
  • For MCP tools, labels for some kinds of server, such as a hosted Chrome server, an artifacts server, or one linked to a plugin.

Tool errors are also now sorted into interrupted or cancelled.

Why

This gives hook steps a finer view of which agent acted and what kind of MCP server it used.

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 doubtIt is not stated whether these new details reach hooks you configure yourself.

See this entry in the whole of v2.1.323 →

Feedback