[FEATURE] Auto-suggest effort level increase for complex tasks

Resolved 💬 2 comments Opened Mar 10, 2026 by m-poh Closed Mar 10, 2026

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

When running complex, long-running tasks (e.g. large refactors, architectural changes, multi-file migrations), users on medium effort may be unknowingly leaving quality on the table — without any signal from the model that the task warrants deeper reasoning.

Currently, adaptive thinking already decides internally whether to think deeply or not on a per-token basis. But this signal never surfaces to the user at the session level. There's no moment where Claude says: "this looks complex — want me to switch to high effort?"

Proposed Solution

Allow Claude to optionally prompt the user to escalate effort level when it detects that the current task is likely to benefit from deeper reasoning. For example:

"This refactor spans multiple modules and involves architectural decisions. You're currently on medium effort — switching to high would give me more reasoning depth for this. Want me to switch?"

The user could confirm or dismiss, and the preference could optionally be remembered for the session.

Alternative Solutions

_No response_

Priority

High - Significant impact on productivity

Feature Category

CLI commands and flags

Use Case Example

_No response_

Additional Context

_No response_

View original on GitHub ↗

This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗