# Features, Berth AI SRE dashboard for Kubernetes

> A read-only AI SRE agent grounded on your live cluster, bring-your-own-model with spend guardrails, and a complete dashboard for health, capacity, endpoints and safe changes.

## AI SRE agent, an assistant you can check against the evidence

The assistant starts with a cluster snapshot and gathers evidence through permitted read-only tools. Coverage depends on Kubernetes permissions and resource support. Raw logs and descriptions require an explicit operator opt-in.

- Single points of failure across Deployments, StatefulSets and operator-managed databases, where resource support permits
- Drain safety: which PDBs would block, and whether other pools absorb the load
- Right-sizing per container and per node pool, with the binding constraint named
- Crash-loop debugging from the previous container's logs plus a runbook
- One conversation that follows you across pages, grounded on what you're viewing

## Bring your own model, your model, your key, your ledger

Connect your own Ollama server, Claude via the Anthropic API, or Claude via Amazon Bedrock. Berth calls the provider directly from inside the cluster. Nothing is proxied, metered or billed by us.

- Daily and monthly token budgets, request rate limit, concurrency cap
- SQLite usage ledger: who asked, which model, tokens, tools, duration, estimated cost
- Cloud providers only behind dashboard auth, and only on Enterprise
- Keys are write-only in the UI and stored with 0600 permissions on your volume

## Operational overview, are my apps up? Is anything broken?

Exposed-endpoint health is synthesised from hostname → DNS → TLS → gateway → route → service. Below it: node and pod readiness, a provisioning verdict, and the warnings that need attention. Gateway API and Ingress, including cross-namespace attachments. "Ask AI" on every screen.

## Capacity planning, which resource is actually binding?

Per node pool: allocatable vs. reserved vs. used for CPU, memory and pod slots, plus the classification (reservation-bound, request-overprovisioned, pod-density-bound). Pods with no requests, PDB count, unscheduled pods. The same numbers the agent uses.

## Operate, ship the next change safely

Cluster-aware builders read the live cluster. The expose-an-app flow creates Deployment → Service → Certificate → Gateway → HTTPRoute in dependency order and verifies the URL is serving. Weighted HTTPRoute backends for canary and blue/green. Human-initiated writes only, gated by an offline-verified license key.

## Self-hosted, by design

Runs in your cluster as one container or binary. Offline Ed25519 license keys; works air-gapped. Free up to 10 nodes, connect your own Ollama server. No SSO, RBAC or multi-cluster yet, on the roadmap.

- [Download Berth](https://berth.agrohi.com/download/)
- [How the AI works](https://berth.agrohi.com/how-ai-works/)