#Bug: Active memory disabled but its output still be persists when verbose is enabled (TG)
6 messages · Page 1 of 1 (latest)
Let me verify this against the actual docs/code so I don’t guess.
Yeah, that looks like a real bug.
I checked the current docs/code path, and the 🧩 Active Memory: ... line is supposed to be a verbose follow-up diagnostic only when active memory is actually running. What seems to be happening instead is that the old active-memory debug line is sticking around in the session state and Telegram verbose mode keeps replaying it.
Quick workaround for now:
- send
/active-memory off - then
/newor/reset
That should clear the stale session diagnostic while keeping /verbose on.
Can you paste:
openclaw --version
If you’re already on latest and it still happens, I’d treat this as a bug worth filing here:
https://github.com/openclaw/openclaw/issues