Roadmap — By stage

From visibility to autonomy.

KubeBolt, your Kubernetes operations platform, advances in stages: first you see, then you act with approval, finally the platform acts on its own under your policy. Windows are indicative: the stage marks when each capability arrives, not the plan that includes it.

The full board

Five pillars, four stages. Each milestone shows what that pillar delivers at that stage, with the detail of how it works.

Stage 01 Free Today
Stage 02 Team Today
Stage 03 Business Q4 2026 – Q1 2027
Stage 04 Enterprise 2027
Kobi
Autopilot
Security
Lifecycle
Fleet
  • Available today
  • Partially available
  • Still building
  • Designed

Tap a milestone to jump to its detail. Autopilot is in open beta on every Cloud plan. A milestone marks when a capability lands, not which plan lets you use it.

Stage 01 · Today

See and resolve

debuts in Free

Available today

Kobi

  • Kobi Copilot — assisted incident resolution
  • kubectl operations, Helm and MCP via agent

Kobi Copilot is Kobi's assisted mode, already working in your cluster: it investigates the incident with real context (events, logs, metrics), proposes the diagnosis and executes actions under your supervision. Every action is audited — who, what and why. It doesn't replace your judgment; it saves you the slow part of the investigation. The agent adds kubectl operations, visibility into your Helm releases and a read-only MCP server.

Autopilot

  • Autopilot in open beta

Autonomous remediation debuts as an open beta on every Cloud plan: you enable it per organization, define the time window it can act in and the mode (suggest-only or execute). Destructive actions require human approval by platform policy, and its usage is metered separately against the same plan credit pool Kobi draws from. Anything awaiting your approval goes to a queue, and every intervention leaves an automatic postmortem.

Security

  • Insights Engine — 24 rules with memory

24 rules that evaluate reliability and configuration from minute one, without writing a single one: missing probes, resource limits, single replicas, unpinned images. Findings feed Kobi, so the copilot already knows the weak spots when an incident hits. Each finding is an episode with its evidence and history: if it comes back it counts as recurring, and bursts group what broke together from one cause. With Slack, Discord or email configured, it alerts you when one fires.

Lifecycle

  • Cloud costs on OpenCost, rightsizing in money

Pulled forward from stage 02: the Cost tab appears on every cluster once OpenCost data flows, without building your own FinOps stack. Monthly run-rate, idle cost, breakdown by namespace, workload and node, and money-priced rightsizing recommendations from your real consumption — with three connection modes in the wizard.

Fleet

  • Per-cluster dashboards with metric history
  • Topology and traffic map

Every cluster has its dashboard: on top, a summary that answers whether anything is on fire; below, the detail, with live metrics and their history from the agent, Metrics Server or your own Prometheus. The topology draws how your resources relate, and the traffic map, which pods talk to each other and to which external destinations.

Stage 02 · Today

Total visibility

debuts in Team

Available today

Kobi

  • Cloud connectors — Kobi reads your AWS account

Many causes aren't in the cluster: a security group, a database, a change in the account. Kobi and Autopilot go out and read your AWS account, read-only, with an IAM role you create and an External ID KubeBolt generates. On Cloud, from the Team plan; on open source it can reach Azure too.

Security

  • Security & Compliance in a single view
  • Kobi over your security findings

Your security tools already produce findings; the problem is they live in four places. KubeBolt detects them (if Trivy, Falco or Kyverno run in your cluster, it finds them on its own), normalizes their findings and presents them in a single dashboard: vulnerabilities, runtime threats, CIS benchmark and policy violations, prioritized. And you can ask Kobi about them like you'd ask a teammate: which CVEs a workload carries, which threats are active, which policies are being violated. Answers come from your real findings, with the detail re-read live from the scanner, not from generic documentation.

Fleet

  • Fleet view and global search
  • ArgoCD apps with their sync and health

Stop hopping between kubectl contexts: every cluster in one panel, with a global search that finds a pod, a deployment or a configmap wherever it lives. Home opens with the shift report: what happened across your clusters while you were away, and what needs you today. And if you run ArgoCD, KubeBolt reads your apps with their sync and health.

Up to here, today.

What follows is in progress or designed. Windows are indicative.

Stage 03 · Q4 2026 – Q1 2027

Act with your approval

debuts in Business

In progress

Usage-based billing starts in this stage, and we'll announce the date in advance; until then, only the plan fee is charged. And the app arrives in Spanish, in beta first.

Kobi

  • Root-cause resolution — the fix reaches your repo
  • Deterministic Skills — fewer credits per incident
  • Cloud connectors for GCP and Azure
  • Kobi remembers every incident it diagnosed

The jump from fixing the symptom to fixing the cause: Kobi applies the fix with your approval and leaves a pull request in the workload's source repo, with the rationale and the before/after in the body, so the next deploy doesn't undo it. Deterministic Skills solve the most frequent diagnostics in code: instant answers, fewer credits. And cloud connectors reach GCP and Azure on Cloud, read-only, as AWS does today. And Kobi remembers what it already knows about each incident: Autopilot doesn't re-read what it already read, and Copilot sees the diagnosis Autopilot left.

Autopilot

  • What awaits your decision, right on Home

Autopilot asks for your approval on anything sensitive and leaves recommendations. When you open Home, "Needs your attention" gathers what only a person can decide: pending approvals, incidents waiting on someone and recommendations not yet applied. Each row opens the incident in its cluster, and it leaves the list on its own once the problem clears.

Lifecycle

  • Rightsizing plans you review and export
  • Actionable costs — orphaned resources

The rightsizing recommendations already exist, priced in money; here they become a plan. Kobi orders the changes by risk, with a rollback for each step and the reason for every workload it leaves out, and every figure comes from the engine, on measured data. You export it as kubectl commands or as a YAML diff, and afterwards Kobi checks that the cluster ended up the way the plan said. And the Cost tab no longer just shows the spend: it points out the volumes nobody mounts and the load balancers with no backends that are still billing, with their cost, and Kobi deletes them with your approval.

Fleet

  • Cluster Timeline — every workload's history

By default, Kubernetes keeps its events for an hour. Cluster Timeline rebuilds the history from what KubeBolt already records: revisions, episodes, bursts, audited actions and metrics, on one timeline per workload and one for the whole cluster. Kobi reads it to tie a failure to what changed before it, as in “started 9 minutes after deploying api v2.14”.

Stage 04 · 2027

Autonomy under policy

debuts in Enterprise

Designed

Kobi

  • Autonomous remediation PRs, with their evidence
  • Kobi Workspace — all your cases in one place

Autopilot detects, investigates and opens the remediation pull request on its own, with the incident evidence in the PR body. Policy decides how much autonomy: from propose-only to auto-merge when your checks are green, always respecting your branch protection. And the Kobi Workspace gathers all your cases, whoever started them: conversations, investigations, plans and incidents, each with its case file and report, and the same cycle of proposed, approved, executed and verified. It also keeps the cluster's memory, with facts Kobi proposes and your team confirms, and a case can be shared or handed to another agent over MCP.

Autopilot

  • Cross-cluster Autopilot — one incident, one thread

The final form: Autopilot operates your whole fleet under a policy you define, with an approval queue for the sensitive and an automatic postmortem for every intervention. You govern the policy; it runs the day-to-day. You give it mandates per cluster and per domain, such as security or FinOps, with a trust ladder that widens what it may do on its own.

Security

  • Autonomous security and FinOps under policy

Security Skills start running on their own under policy: CVE triage by real exposure, runtime threat response and policy drift correction. Your team reviews the exceptional, not the routine.

Lifecycle

  • ROI dashboard — savings measured, not estimated
  • Scheduled non-production shutdown

A history of every action Kobi and Autopilot apply, with the savings measured after applying it rather than estimated, exportable to CSV for your FinOps report. And scheduled shutdown of the environments nobody uses at night: they scale to zero at the hour you pick, per namespace or per cluster, and come back when it's time.

Fleet

  • Multi-cloud cockpit with reconciled cost

A single cockpit for AWS, GCP and Azure: findings, cost and insights aggregated cross-cluster, and cost reconciled against each cloud's real billing.

Want to know when something goes live?

What's next

One note per release, no noise.

Missing something? Tell us — it shapes this roadmap