From the first signal to a change you can explain.

See broker availability, offline partitions, under-replication, and rack findings together. Source times and missing evidence stay visible.

Compare sampled activity, stored replica bytes, and leader distribution. Open a broker to inspect the partitions behind the imbalance.

Inspect group state, members, partition lag, and lag trends. Review exact offsets and retained ranges before a licensed offset reset.

Estimate partition activity from watermark changes and log-dir sizes. No JMX setup, broker agents, or message consumption required for monitoring.

Inspect individual messages with filters, jump to an offset or timestamp, or tail a topic live. Monitoring itself never consumes.

Export cluster health, sampled activity, consumer lag, and execution metrics to Prometheus for your dashboards and alerts.

Inspect current and proposed placement, reasons, and projected balance. Rack-aware proposals refresh as the cluster changes.

The same cluster snapshot and settings produce the same plan. Simulate broker or rack loss, added capacity, and traffic changes without changing Kafka.

Apply reviewed plans with live health checks, adjustable replication throttles, and per-broker concurrency limits. Hard health blockers prevent submission.

Moves are picked so ISR stays healthy while data is in flight.

Track queued, active, completed, and failed assignments. Adjust throttles or request cancellation of remaining work, then inspect the resulting placement.

Drain or decommission a broker with RF preservation and rack-collapse reporting; move log dirs between disks.

Run critical repairs, replication-factor repairs, and activity balancing within configured windows. Inspect schedules, cooldowns, and recent attempts in Automation.

When balance is structurally unreachable, Pilot names the metric it cannot fix and stops.

Find recorded requests by actor or action and inspect their results. Supported reversals check current state first; reassignment review shows what can be restored and what is blocked.

OAuth2/OIDC SSO via Entra ID, Google, GitHub, or Keycloak, plus revocable read- or write-scoped access tokens.

AI- and MCP-initiated changes wait for a distinct approver. Hard health blockers cannot be overridden.

Restrict sign-in by SSO group, organization, or domain. Access tokens carry read or write scope.

Pure Go, CGO-free, on a distroless/static image. No external database - state lives on compacted Kafka topics.

Ed25519-signed licenses validate offline. No phone-home, air-gap friendly.

Deploy in minutes with a bootstrap-servers address - as a container, a Compose stack, or on Kubernetes.

Add host metrics and broker logs with optional mTLS agents. Inspect connected agents and host telemetry, with rolling restarts where Pilot manages the broker lifecycle.

Work in the UI, automate through the REST API, or connect MCP tools. The optional assistant uses your chosen LLM provider; cluster context is sent to that provider.

SASL PLAIN, SCRAM, Kerberos, and OAuth over TLS or mTLS - no broker-side changes needed.