Follow Discord
Sweep 03 Nov 2026 · 22:27Z Build v2.1.318 536 read Stable v2.1.311 Latest v2.1.318 Next v2.1.318 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.317 ·

New sandbox status report for host apps, including where locking policies come from

Claude Code can report its sandbox state to a host app, including support, locks, restrictions and which managed-settings sources apply

Group of 2 You'll notice No documentation found Improvements
JSON All of v2.1.317
You'll noticeTier: how much it should matter to you
3Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
SandboxArea: what it touches
ImprovementsKind: in v2.1.317,
ImprovementsSection of the release

What

The sandbox is the protective boundary that limits which files and network addresses commands run by Claude Code can reach. Claude Code now has a function that reports the sandbox's status to a host, meaning an app that embeds Claude Code, such as an editor plugin or a program built on the SDK. The report contains:

  • supported and unsupported_reason: whether the sandbox works here, and why not if it does not.
  • locked: whether the sandbox settings can be changed.
  • policy_origins: an optional list of the managed-settings sources that apply. Managed settings are settings an organisation pushes to its users. The codes include remote, plist, hklm, file, parent, helper, remote_unknown and unreadable.
  • mode.
  • restrictions, with lists for files (fs) and network.
  • dependencies.

Why

A host app can show what the sandbox is doing without guessing. With policy_origins it can also tell you where the policy that locks your sandbox settings came from, for example a remote policy, a macOS plist or a Windows registry key.

Read from
Names in the bundlepolicy_origins

See this entry in the whole of v2.1.317 →

Feedback