Follow Discord
Sweep 22 Oct 2026 · 21:04Z Build v2.1.309 528 read Stable v2.1.308 Latest v2.1.309 Next v2.1.309 Feeds RSS JSON llms.txt llms-full.txt Unofficial
HomeReleasesv2.1.309entry #82

Claude Code v2.1.309 ·

Claude Code now reads the openWorldHint that MCP tools declare

Claude Code now passes along the openWorldHint an MCP server sets on a tool, as well as its read-only and destructive hints

You'll notice Improvements
JSON All of v2.1.309
You'll noticeTier: how much it should matter to you
2Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
MCPArea: what it touches
ImprovementsKind: in v2.1.309,
ImprovementsSection of the release

Unclear What Claude Code does with openWorldHint once it reads it is not stated.

What

MCP servers are outside services that give Claude extra tools. A server can attach hints to each tool describing how it behaves. Claude Code already read the readOnlyHint and destructiveHint hints. It now also passes along openWorldHint.

The information Claude Code's automatic permission classifier receives about a tool is now built from that tool's own name. The classifier is the part that decides whether an action can go ahead without asking you.

Why

The extra hint can feed into permission decisions, including the automatic ones.

Read from
Names in the bundleopenWorldHint
What the documentation says
Documented inclaude-code/agent-sdk/custom-tools
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 doubtWhat Claude Code does with `openWorldHint` once it reads it is not stated.

See this entry in the whole of v2.1.309 →

Feedback