Unclear It is not known what other word can replace auto or when, so it is unclear what to type in those setups.
What
The /effort command sets how much effort the model spends on its answers. Several things around it changed.
- The check for
defaultorautonow looks at the trimmed text you typed, plus one extra check. Before, it tested a variable that was never set, which looks like a bug. - The usage text, suggestions and the tip about switching away from
maxnow get the auto keyword from a function instead of the fixed wordauto, so the word can differ. The default option reads "Use the default effort level for your model". - When a reply starts, the session now records the effort that was sent and the effort for the turn, from
sentEffort,turnEffortandeffort.
Why
/effort default and /effort auto may now work where they did not before. The name of the auto option can vary, so check the usage text if it looks different.
Names in the bundle/effort
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 since
A small documentation edit on anthropic-managed touched a line naming /effort after this was published.
* [Effort](https://platform.claude.com/docs/en/build-with-claude/effort)build-with-claude/claude-models-on-google-cloud/anthropic-managed see the edit
Confirmed since
Anthropic's documentation has since written up /effort, on Model configuration.
When Claude Code sends no effort level to a [supported model](#adjust-effort-level), the [API's default](https://platform.claude.com/docs/en/build-with-claude/effort) applies instead of Claude Code's. To have Claude Code send none, start `…model-config see the edit
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubt
It is not known what other word can replace `auto` or when, so it is unclear what to type in those setups.
Anthropic's documentation agrees
Anthropic's documentation has since written up /effort, on Model configuration.
Anthropic's release notes agree
Fixed the /model and /effort cache warning saying the re-read waits for your next message; a change made while Claude is working can trigger…