Follow Discord
Sweep 09 Oct 2026 · 17:27Z Build v2.1.296 517 read Stable v2.1.287 Latest v2.1.296 Next v2.1.296 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.296 ·

Deny rules can match extra names through denyMatchNames

A new denyMatchNames list adds extra names for deny-rule matching on prompt-type commands and feeds reserved labels for MCP-loaded tools

Group of 2 You'll notice Improvements
JSON All of v2.1.296
You'll noticeTier: how much it should matter to you
2Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
PermissionsArea: what it touches
ImprovementsKind: in v2.1.296,
ImprovementsSection of the release

Unclear Which commands get these extra names, and what the names are, is not known.

What

A new denyMatchNames list gives commands and tools extra names to match against. A deny rule is a permission setting that blocks something by name.

  • Prompt-type commands now expose a denyMatchNames list. These are commands whose content is a prompt handed to Claude, which can include skills. The list joins the names already used for matching: aliases, shedAliases and unqualifiedName. A second helper that builds the list of match names also includes it.
  • Tools loaded from MCP now get a reservedLabels list. MCP (Model Context Protocol) is a standard way to connect Claude Code to outside tools and services. The list is built from denyMatchNames entries that contain a : and match the names of other known tools. MCP-loaded tools had no such list before.

Why

In practice, a command, skill or tool can now be matched under more names than its usual name and aliases. Name matching is what deny rules rely on. If you write deny rules by name, keep in mind that a command or tool may now answer to extra names it did not answer to before.

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 doubtWhich commands get these extra names, and what the names are, is not known.

See this entry in the whole of v2.1.296 →

Feedback