Problem

The slash-commands documentation states:

/new [model] starts a new session; /reset is the reset alias.

This is misleading and factually incorrect. They are not aliases. They behave completely differently:

• /new → Creates a fresh session ID. The old session is archived and still accessible.
• /reset → Literally resets the current session in place. The old context is gone. No coming back.

Calling /reset an "alias" of /new is a documentation error that confuses users into thinking they can use either interchangeably — until they lose their session data.

The docs contradict themselves

Further down the same document:

Control UI intercepts typed /new to create and switch to a fresh dashboard session; typed /reset still runs the Gateway's in-place reset.

So the docs themselves admit they behave differently. Yet the summary line still sells them as equivalent aliases. Make it make sense.

This is a regression that was apparently endorsed by a maintainer comment in #53335:

Docs describe /new as session reset behavior: The slash-command docs state /new [model] starts a new session and /reset is the reset alias, matching the fixed behavior.

That comment uses the same faulty phrasing to justify closing a bug. A whole issue closed based on a documentation sentence that perpetuates the confusion rather than clarifying it.

Requirements

1. Remove "/reset is the reset alias" — that sentence has no place in accurate documentation
2. Clearly distinguish the two commands: /new = new session (old one archived); /reset = wipe current session clean
3. Update all occurrences of this phrasing across the docs — don't leave another landmine
4. Consider renaming /reset to something more explicit (e.g., /clear, /wipe), or at minimum add a warning about data loss

Environment

• OpenClaw: 2026.5.2 (8b2a6e5)
• Platform: macOS / webchat / TUI — docs issue, not platform-specific

A project that prides itself on "right answer, delivered cleanly" should start by getting its own command semantics straight in the docs.