Support a plain-language argument to the /model skill for faster, non-interactive switching

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

_To accelerate the rate of a single Claude session_, it is often helpful to provide active followups to a pending operation. It is helpful to say that pending operations almost always have potential followup. With good planning, followup actions benefit from using a better model, which is well understood.

A Sonnet session leading to Opus is a pattern that plays out millions of times per day.

Fact: power users run big sessions with a lot of queued messages in a single session. But switching models is a blocking action and can't be set inside of a queue to run automatically.

Proposed Solution

A way of improving this relationship is if the /model skill took an optional argument. Proposal is that this optional argument have plain verbosity, i.e. "Opus" instead of "Opus 4.7" which would default to the highest version available. It's worth nothing this is compatible with the verbiage of the current selection dialog. So in other words, if we could say /model Opus as a queued message, the user would never be forced to wait when she needs extra high effort next. Conversely, she could segue into Sonnet more quickly, increasing the speed of the overall workflow, and empowering power users who don't mind typing an extra few characters.

Alternative Solutions

An alternative to allowing an argument to the /model skill would be to make the interactive mode of the skill non-blocking. You'd still be required to take an action like in the current interface, but the contract would be sealed when the action finishes, without affecting the active model of the running process.

Priority

High - Significant impact on productivity

Feature Category

Interactive mode (TUI)

Use Case Example

_No response_

Additional Context

_No response_