A skill is a packaged set of instructions Claude can follow for a particular job. Skills begin with a short header of settings called frontmatter. When a server-controlled switch is on, the compatibility field from that header is added to the skill's text that Claude reads. The added text has a length limit.
It is followed by guidance for Claude. If the skill needs your own computer or a private network and nothing in the current environment can reach it, Claude should say so and offer what it can still do.
The switch is off unless it is turned on remotely, and it has read off for this site's account.
This changes how skills that depend on local machines or private networks behave in cloud sessions: Claude explains the limit instead of failing quietly.
tengu_fancy_canyon Off in both readingsThe flag server returned off for the account this site reads and for the anonymous baseline. A reading of off cannot rule out a rollout these two readings sit outside of.
This account: off · anonymous baseline: off · compiled default in v2.1.321: off
These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.
Read once, for one account on one subscription tier, against v2.1.321. It isn't a statement about your account. What a flag value here can and cannot tell you
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.
2. **1M context window is the default:** Claude Opus 5.5 serves the full 1M token [context window](https://platform.claude.com/docs/en/build-with-claude/context-windows) by default. If your client passes a context-window beta header for co…models/opus-5-5/migration-guide see the edit
Anthropic's documentation has since written up compatibility, on migration-guide.
New in this build: tengu_fancy_canyon