Open WebUI v0.12.4 adds on-demand skill loading and says it contains unlisted security fixes
Open WebUI v0.12.4 adds optional on-demand loading of skills and shows when chat context is compacted; its release notes say the release contains security fixes that are not itemised and recommend updating.
- First seen
- (via GitHub releases feed (watcher seed run))
- Source published
- (we saw it 19.7 hours later)
- Brief published
What changed
- Security: the notes say the release contains security fixes that are not itemised in the changelog, recommend updating all deployments as soon as possible, and promise advisories after CVE assignment.
- Skills: a new admin option, "Load selected skills on demand" (
ENABLE_SKILL_LAZY_LOADING), lists selected skills for the model to load when needed instead of sending their full instructions with every message. It is off by default and needs the built-in Skills tool. - Context compaction now shows a "Context compacted" line where older messages were summarized, and also runs in temporary chats and in short chats that are already over the threshold.
- Administrators can hide individual controls from users under Admin Settings > Interface or with
UI_VISIBILITY; administrators still see every control. - Performance: folder changes, channel posts and the admin analytics dashboard are described as faster on busy instances, and tools from tool servers now reach the model in the same form after a restart so provider prompt caches keep working.
In the source’s words
“We strongly recommend updating all deployments as soon as possible to have the latest security fixes in place.”
Early verdict
Self-hosters should plan the update soon; the vendor itself recommends it.
Basis: Based only on the release notes' own security statement; the fixes and advisories were not available to read and we have not run the release.
What we have not verified
- We did not run this release or test any of the new options.
- The security fixes are not listed in the release notes, so we cannot say what they cover or how severe they are.
- Performance claims are the project's own wording; we did not measure them.
This brief is Reported: written from the sources below, not from running the release ourselves.
Sources
- Open WebUI v0.12.4 release notes (GitHub) (primary source) (fetched 2026-10-26)
Related
Spotted an error? See our corrections policy or email the address on the contact page. Markdown version for agents: /news/open-webui-v0-12-4-skills-compaction-security.md
Want these briefs in your Google Top Stories? Add Agentic Weekly as a preferred source on Google.
Free · weekly · about six minutes
Get The Weekly Verdict, free
What shipped this week, what is worth switching to, and what to ignore. We run what we can and label what we only read.
- One email a week. Each release gets one call: update now, worth a look, or skip.
- Every claim shows its evidence. Reported means we read the sources; Tested means we ran it, with the exact scope and the failures left in.
- A free AI Tools Cheat Sheet (web page and printable PDF) lands in your inbox when you confirm.
Free. Unsubscribe anytime. No spam. We email a confirmation link first and add you only after you click it. We use your email only to send this newsletter and never sell it. Privacy