Upgrade Copilot

Built for Kubernetes, Add-ons, Application Services and 100s of Open Source Software Projects.

Automate every phase of your upgrade cycle — from parsing release notes and diffing Helm charts to mapping CRD, dependency, and config changes. Our agent flags breaking changes, EOL risks, and critical defects early and identifies incompatible components and API deprecations. Provides pre‑/in‑/post‑flight validation checks, delivers ready‑to‑apply IaC automation code, and much more …

What’s Included:

Upgrade Assessments: Rapid, context-aware reports that surface all upgrade blockers, risks, and version recommendations for your cluster, add-ons and OSS projects.
Upgrade Templates: AI-curated, reusable workflows that encode the safest upgrade path for any cluster or add-on or OSS project.
Upgrade Plans: Context-aware execution blueprints instantiated from a Template, tailored to an environment and ready to run.

Start for free Book a demo

Chkk dashboard highlighting the eks-prod-1 cluster

AI-curated, Production-Safe Upgrade Workflows

Generates Upgrade Assessments, Templates & Plans for every cluster via the Knowledge Graph + RSig DB, so you start from a ready-made, environment-specific workflow instead of a blank page.
Live workflow tracker lets teams assign owners, set actions, add comments/notes, mark complete or skip, and preserves a full audit trail—plus diff-based PRs for Terraform, Helm, or Kustomize.
Keeps your fleet current without forced upgrades or premium-support surcharges.

Focused view of Chkk's Upgrade Copilot, focused on the Add-On Changes step of Pre-work.

Contextualized Breaking Changes & Smart Version Guidance

Collective Learning distills upstream release notes, GitHub issues, and docs into source-linked change logs that apply only to the versions you run—no more reading through dozens of pages.
Copilot calculates the safest next hop by weighing Kubernetes compatibility, add-on matrices, EOL timelines, and real-world mileage—never “latest for latest’s sake.”
Identifies multi-hop constraints (e.g., Consul 1.10 → 1.12 → 1.14 → 1.16 → 1.18) and assembles the exact safe path for each stateful component.
Differentiates required, recommended, and optional upgrades and links to authoritative upstream notes for every suggestion.
Proactively warns when running versions edge toward end-of-life or surcharge windows, giving teams clear action deadlines.

Find Out Application Dependencies & Unknown Incompatibilities

Auto-maps every add-on, OSS project, CRD, custom resource, node group, and workload to surface hidden add-on → application couplings and upgrade blockers.
Flags misconfigured PDBs, deprecated or removed APIs, and webhook chains that could stall rolling upgrades—before code reaches staging.
Recommends the exact AMI / kernel / container-runtime versions proven compatible with your target Kubernetes release and every CNI or Service Mesh you run.
Warns when cgroup v1 → v2, syscall, or kernel shifts might break CNIs or service meshes, and recommends patched images where needed.

Helm, CRD, & RBAC Diffs You Can Act On

Inline, annotated diffs show exactly what will change in Helm values, Kustomize overlays, Argo-CD app-sets, or raw YAML, eliminating manual side-by-side reviews and silent drift.
Compares old vs. new CRD schemas and every stored Custom Resource, guiding safe migrations and avoiding validation errors.
Audits RBAC additions or removals so required permissions evolve with every upgrade.

Add-On & Multi-Hop Paths

Comprehensive Pre-flight, In-flight, & Post-flight Checks

Auto-attaches pre-flight, in-flight, and post-flight readiness checks for each Kubernetes and add-on version; results stream into Copilot’s workflow tracker.
Lets you embed custom checks alongside the library for bespoke assurance and compliance gates.
Early failure detection slashes rollback time and protects SLAs during maintenance windows.