🦊 GitLab

AI adoption can stall when costs aren't predictable. GitLab 18.11 introduces spending caps and per-user budget controls for GitLab Credits, so you can scale GitLab Duo Agent Platform. Set a ceiling. Roll out to more teams. Ship with confidence. Learn more. about.gitlab.com/blog/gitlab-18…

AI made writing code faster. Setting up the pipeline to run it and understanding how delivery is actually going? Still on you... until now. CI Expert (Beta) generates a working pipeline config from your repo in minutes, and Data Analyst (GA) answers natural-language questions from live GitLab data in Agentic Chat. Both run inside the same platform as your code, pipelines, and issues. about.gitlab.com/blog/ci-expert…

Read the full 18.11 release notes:docs.gitlab.com/releases/18/gi… 🚀

Faster code generation hasn't meant faster delivery. GitLab 18.11 is here to address that gap. ⚡️ Agentic AI now covers pipelines, security, and delivery analytics so your entire delivery chain moves as fast as your IDE. ✨ What's new in 18.11: ➡️ Automate Remediation: Agentic SAST Vulnerability Resolution is now GA, generating ready-to-merge code fixes before vulnerabilities reach production. ➡️ Two New Agents: The CI Expert Agent (beta) can spin up a working pipeline in minutes, and the Data Analyst Agent (GA) answers live delivery questions in natural language. ➡️ Spend Controls: New subscription-level and per-user caps for GitLab Credits give organizations control over on-demand AI spend at scale. ...and more improvements!

Ready to dive deeper into 18.11? Learn more here: Learn more about what's available in 18.11 here: about.gitlab.com/blog/automate-… about.gitlab.com/blog/ci-expert… about.gitlab.com/blog/gitlab-18…

Learn more about the 18.11 release: Learn more about what's available in 18.11 here: about.gitlab.com/releases/whats…

Faster code generation hasn't meant faster delivery. GitLab 18.11 helps address that gap with agentic AI across pipelines, security, and delivery analytics. Here are the key changes in this release🧵:

Claude Opus 4.7 from @AnthropicAI is now available in GitLab Duo Agent Platform. Teams get stronger reasoning and more precise agents across every workflow. about.gitlab.com/blog/claude-op…

GitLab Duo Agent Platform agents can now call foundation models through Vertex AI, including Gemini. Every agent action is governed by the same controls as your human developers, and you can choose to have usage count toward your existing Google Cloud commitments.

Building an AI agent is the easy part. Securing it and deploying it is where many teams stall. Our own Regnard Raquedan will be at @googlecloud Next, showing how GitLab + Agent Engine takes agentic development all the way from build to ship. Join us April 22-24, booth #2324.

We answered community questions during the last episode of The Developer Show. Here’s our take on how and when to use multi-agents. 👇

Learn more: docs.gitlab.com/user/duo_agent…

The attack surface grows every time an AI agent commits code. Most security tools scan after the fact. GitLab Duo Agent Platform’s Security Analyst Agent works inside the workflow, triaging vulnerabilities, filtering false positives, and generating fixes before anything ships.

#SoftwareDevelopment coding with AI tools + platforms, including AI agents, can accelerate coding, but lead to bottlenecks downstream in s/w delivery. These bottlenecks result from fragmented context. -- Bill Staples, CEO @GitLab cxotalk.com/episode/intell… #CXOTalk #AI #AIagents

Learn more here: about.gitlab.com/blog/gitlab-du…

GitLab Duo CLI is now in beta. Bring GitLab Duo Agent Platform to your terminal for local development workflows. Pull context from issues and Agentic Chat sessions, iterate in plan mode, then let Duo implement, commit, and open MRs.

Read more: about.gitlab.com/blog/gitlab-an…

GitLab Duo Agent Platform agents can now call foundation models through Vertex AI, including Gemini. AI agents that build software need more than a capable model. They need context from the issue that started the work, the repository it touches, the pipeline that validates the result, and the security policies that govern what ships.