Reported News
LobeHub v2.3.0 turns on Proactive Bot by default: bots now decide for themselves when to speak in channels
LobeHub v2.3.0, released Nov 1, 2026 with 145 merged PRs, turns Proactive Bot on for every bot; the release notes say "Bots stop waiting to be @-mentioned." Bots judge each channel message, log every decision in a new Wake records tab, and can be switched off per channel.
- Changes reported
- 5
- listed below, each from a source we read
- We tested
- 0
- we did not run it ourselves
- Gaps we flag
- 3
- things we could not verify
- Sources
- 1
- 1 primary, linked below
What changed
| # | What the source says | Status |
|---|---|---|
| 01 | Proactive Bot, previewed behind a rollout flag in v2.2.21, is now on for every bot: the bot judges each message in the channels it watches and decides whether to speak up, stay quiet or wait. Owners can turn it off in the new Proactive participation block of the channel Wake policy tab. | Reported |
| 02 | Bots with owner-written duties no longer wait for a message that falls inside them, and they also step in on open questions nobody has answered. The release notes say bots still stay out of replies addressed to someone else and that this adds no extra model call. | Reported |
| 03 | A new Wake records tab lists every message a bot judged, with a 30-day summary (woke, skipped, deferred, proactive), the model score and a readable reason per message, and lets owners mark each decision right or wrong. | Reported |
| 04 | Other highlights: real-time multi-person page editing with page Agent edits run in a sandbox, threaded Agent replies in channels, memory consolidated into contexts and dossiers shown as stories, declarative dashboards with a createDashboard tool, and the persistent sandbox out of preview. | Reported |
| 05 | Self-hosted installs get six database migrations (0193 to 0198) that run on startup. The notes warn that 0196 builds indexes on the user_memories_* tables inside a transaction, blocking writes while it runs, and suggest creating them with CREATE INDEX CONCURRENTLY first if those tables are large. | Reported |
reported by the sourcemeasured or run by usnot verified
Primary source: LobeHub v2.3.0 release notes (GitHub). We are the fast, independent index; the vendor’s own page is the source of record for this release.
In the source’s words
“Bots stop waiting to be @-mentioned. With Proactive Bot on by default, every bot reads its channels, steps in when it can help, and shows its owner every call it made.”
“This adds no extra model call.”
Early verdict
Self-hosters should read the migration note before upgrading, and teams with bots in busy channels should expect them to start speaking unprompted unless they switch Proactive participation off.
Basis: Based only on the v2.3.0 release notes as read on Nov 1, 2026; we have not run it.
What we have not verified
- We did not run this release ourselves, so we have not seen how often proactive bots speak up in practice or how accurate their wake decisions are.
- We did not check the claim that proactive judging adds no extra model call, or measure any cost change.
- We did not run the 0193 to 0198 migrations on a self-hosted install or time the 0196 index build.
This brief is Reported: written from the sources below, not from running the release ourselves.
Sources
- LobeHub v2.3.0 release notes (GitHub)primary source · fetched 2026-11-01
Related
Spotted an error? See our corrections policy or email the address on the contact page. Markdown version for agents: /news/lobehub-v2-3-0-proactive-bots.md
Want these briefs in your Google Top Stories? Add Agentic Weekly as a preferred source on Google.
The Weekly Verdict · free
5 new-model use cases. Every Monday.
A free email every Monday: the week's few big AI releases, then 5 short use cases for the new models, each with a prompt you can paste.
Free. Double opt-in. One click to unsubscribe.