You'll noticeTier: how much it should matter to you
1Useful: my rating, 1 to 5
1Signal: worth watching, 1 to 5
SkillsArea: what it touches
ImprovementsKind: in v2.1.283,
ImprovementsSection of the release
Unclear How removing the separate naming for synced skills changes what you see is not clear.
What
A skill's allowed-tools field can use placeholders that Claude Code fills in with a folder path. What happens when there is no folder to fill in has changed:
CLAUDE_SKILL_DIR: if the skill has no folder of its own, entries that use it are dropped and a warning is logged.
CLAUDE_PROJECT_DIR: if the session has no working directory, entries that use it are dropped.
Before, such placeholders were left unfilled, or the project one fell back to the current directory. Synced skills also no longer get a separate naming scheme of their own.
Why
An unfilled placeholder inside a permission grant could match something it was never meant to. Dropping the entry closes that gap, though a skill that relied on it will now ask for those tools instead.
Read from
Names in the bundleallowed-tools
Since it was published
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
Added sinceA small documentation edit on Extend agents with skills touched a line naming allowed-tools after this was published.In SDK sessions, you can pre-approve tools for a project or personal skill with the skill's [`allowed-tools`](/docs/en/skills#pre-approve-tools-for-a-skill) frontmatter or with the `allowedTools` option (`allowed_tools` in Python) in your …agent-sdk/skillssee the edit
Confirmed sinceAnthropic's documentation has since written up allowed-tools, on Claude Code changelog.* Fixed plugins from marketplaces, claude.ai and npm pre-approving their own tools via `allowed-tools` under managed `allowManagedPermissionRulesOnly`; only plugins from an official Anthropic source or a source that managed settings vouch …changelogsee the edit
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 doubtHow removing the separate naming for synced skills changes what you see is not clear.
Anthropic's documentation agreesAnthropic's documentation has since written up allowed-tools, on Claude Code changelog.
