GitLab 18.10: AI, saved views and faster planning
GitLab 18.10 adds work item saved views, free-tier credits, passkeys and AI-native security workflows for teams modernising delivery.
GitLab 18.10 changes how teams plan and secure work
GitLab 18.10 is one of those releases that affects several parts of the software delivery workflow at once. If you are tracking GitLab 18.10 release notes, the headline is not only new AI capabilities. It is also the arrival of work items list saved views, passwordless sign-in with passkeys, free-tier GitLab Credits for agentic AI, and a much more practical security story for teams that want to reduce manual effort.
Planning becomes less repetitive
The new work items list brings different work item types into one view. Saved views then let teams store their preferred filters and layout choices, so planning sessions do not begin with the same setup steps every time. For product managers, engineering leads and delivery managers, that means less friction when reviewing the same portfolio from different angles.
This is exactly the sort of keyword cluster that keeps surfacing in search data around GitLab 18.8 and GitLab 18.10: people want to know whether GitLab can handle planning, saved views and agile workflows in one place. The answer in 18.10 is clearly yes, and teams can compare that experience with other pages on our site at gitlab.consulting/en-gb.
AI-native security becomes more actionable
The most visible security update is SAST false positive detection using the GitLab Duo Agent Platform. Security teams have long wanted fewer noisy findings and faster triage. GitLab 18.10 moves in that direction by using an LLM-assisted workflow to identify findings that are likely false positives before developers spend time on them.
Equally important, the release connects detection and remediation more closely. That matters because vulnerability management often fails not at detection, but in the handoff from security to development. By bringing remediation into the workflow, GitLab reduces context switching and makes it more likely that teams will actually fix what they find.
Agentic AI without a hard subscription jump
Another important change is commercial rather than technical: free GitLab.com teams can now purchase a monthly commitment of GitLab Credits and use the GitLab Duo Agent Platform without a subscription upgrade. That lowers the barrier for teams that want to test agentic AI in a realistic setting before committing larger budgets.
This is relevant for organisations in Central and Eastern Europe, and for UK teams that are exploring AI adoption carefully. Many buyers are searching for terms like GitLab credits, GitLab Duo Agent Platform and agentic AI, and this release is likely to create more of that demand.
Why this release matters
Taken together, GitLab 18.10 is not just an incremental version bump. It is a signal that GitLab is connecting planning, identity, security and AI into a more coherent operating model. If your team is evaluating whether to modernise planning workflows or reduce vulnerability triage overhead, this release gives you several concrete reasons to look again.
For a fuller comparison of how these capabilities fit into a broader deployment and adoption strategy, the consulting team can help you map them to your own environment and governance model.
Need help adopting GitLab 18.10?
If you need GitLab consulting, enablement or licence procurement support in the Czech Republic, Slovakia, Croatia, Serbia, Slovenia, North Macedonia or the United Kingdom, get in touch with our team and we will help you plan the right next step.
Tags:GitLab 18.10saved viewspasskeysGitLab DuoSASTAI
Other languages:ČeštinaSlovenčinaHrvatskiSrpski (Latinica)