A skill is a set of instructions Claude follows for a particular task. When a server-side switch is on, the Claude API skill gets different instructions about signing in. Claude:
- runs
ant auth statusfirst to see whether you are signed in - if not, offers to sign you in with
ant auth loginfirst, and an API key second - handles setups with no browser, and cloud sessions where
CLAUDE_CODE_REMOTEis set - never asks you to paste codes or keys into the chat
Before, the skill told Claude to ask you for a key when no sign-in was found. The switch is off unless it is switched on remotely, and the flag server returned off both for this site's account and without an account.
It changes what Claude tells people starting with the Claude API about signing in, and keeps keys out of the conversation.
tengu_ancient_axolotl 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 v2.1.319: 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 v2.1.319. It isn't a statement about your account. What a flag value here can and cannot tell you
| `CLAUDE_CODE_REMOTE_MCP_CONFIG` | Path to a file in `CLAUDE_CONFIG_DIR` holding the control plane's delivered MCP servers. If your wrapper rebuilds the child's environment, carry this variable, [`CLAUDE_CODE_REMOTE`](https://code.claude.…see the edit
CLAUDE_CODE_REMOTE on Customize sessions in self-hosted environments