New rule sets for GitHub Actions and Argo CD, a build determinism check for reproducible pipelines, bulk repository import, and per-repository analytics. Explore the full Convergence release →
New features
-
📥
Bulk repository import
Onboard your whole estate in one pass instead of adding repositories one at a time.
- Import sources: Import from a JSON or CSV file, or connect a named integration.
- Guided wizard: A UI wizard walks you through selecting and importing many repositories at once.
- Import filtering: Narrow the candidate list by repository name, branch, language, last activity, owner, and more before you commit.
-
📈
Repository analytics & scan progress
See what's happening, repo by repo, run by run.
- Repository-scoped analytics: Analytics broken down per repository.
- Scan progression indicator: Track a verification run's progress in real time.
- UI overhaul and broader UX improvements: Easier day-to-day verification across the platform.
Rule engine
-
⚙️
GitHub Actions verification & security
A new rule set for workflow verification scanning and security checks in GitHub Actions pipelines.
- New rule set: Dedicated coverage for GitHub Actions workflows.
- Security scanning: Surfaces permissions, secrets usage, and unsafe patterns before production.
-
🚀
Argo CD rule set
A new rule set for GitOps verification and Argo CD best-practice checks.
- New rule set: Validates GitOps configs and Argo CD deployment practices.
- Delivery pipeline coverage: Strengthens security posture across Kubernetes GitOps estates.
-
🔒
Build determinism
A build that resolves different dependency versions on every run can't be trusted or reproduced — and that's a supply-chain risk, not just a quality issue.
- New rule: Flags non-deterministic install commands (e.g.
npm install) in CI configs.
- Deterministic alternative required: Confirms that reproducible, lockfile-respecting commands (e.g.
npm ci) are used instead.
- Drift detection: Reduces exposure to silent dependency drift between builds of the same pipeline.
- Stronger provenance: Strengthens the guarantee that what was scanned is what actually shipped.
Verification expands to AWS and the ML supply chain, with a sharper rule engine and scale for the biggest estates. Explore the full Overworld release →
New features
-
☁️
AWS support
Bring cloud-native AI workloads inside your verification perimeter.
- AWS agent scanning: Detect and review agents running on AWS.
- Guardrails detection: Surface guardrails configured for AWS workloads.
- AWS Bedrock review: Review Bedrock usage across your pipelines.
-
🧠
Model security & MLSecOps
Security for the ML supply chain, built for DevOps teams.
- MLSecOps rules: New rules and MLflow best practices.
- Model security: Model security brought to DevOps teams.
- New technology detections: Broader coverage across the ML stack.
-
✨
UI improvements
A clearer read on every run.
- Readability: UI improvements for better readability.
- Run statuses: New statuses for verification runs.