Claude Code Surfaces Power-User Slash Commands

This reads like a productivity layer upgrade more than a headline launch: Claude Code is accumulating workflow primitives that make long agent sessions easier to steer, recover, and automate.

• –`/less-permission-prompts` looks like a safety/usability valve for trusted commands, reducing the friction that usually pushes users toward full auto mode.
• –`/recap` is the most broadly useful: it formalizes context compression, which matters once sessions get long enough that human memory and agent memory start diverging.
• –`/Advisor` suggests a tiered-model workflow, where Sonnet handles the first pass and a stronger agent steps in only when needed, which is a sensible way to conserve higher-end model spend.
• –`/Dashboard` hints at Claude Code expanding beyond terminal code work into orchestration and operational views, though the Reddit post’s description is still pretty thin.
• –The bigger signal is that slash commands are becoming a control surface for agent behavior, not just shortcuts for common prompts.