AI Agent Hub
Back to plugins
⚙️

dsh-aiops

Workflow Updated 2026.09.08

Run the following command in DeepSeek Harness:

dsh plugin install jarvan1/dsh-aiops

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install jarvan1/dsh-aiops in the DeepSeek Harness terminal to install this plugin; the full source code is available at https://github.com/jarvan1/dsh-aiops

About this plugin

When Alertmanager fires, on-call engineers often bounce between Prometheus, Kubernetes, and logs, stitching context by hand to pin down the root cause. dsh-aiops collapses that chain into a single automated, read-only workflow: an alert comes in, a structured diagnosis comes out, and no production write operation is ever touched.

The core capabilities revolve around an event closed-loop — an authenticated webhook receives Alertmanager pushes, fingerprints route into a durable session queue with resource budgets, alert-time query windows scope dynamic Prometheus calls and Kubernetes read-only API access (no kubectl required), and the result is a structured incident report, append-only operator feedback, and workspace-scoped routing audit logs. Connectivity and minimum RBAC are verified live in the AIOps Portal, and routing policies support versioned atomic saves with full change traceability.

It is built for SRE and platform-engineering teams that already run Alertmanager + Prometheus + Kubernetes and want a one-stop diagnostic experience inside DSH Web. Read-only scoping, no extra credential exposure, and auditable history make it safe to drop into an existing on-call pipeline.

Use Cases

  • Alertmanager push automatically pulls Prometheus metrics and Kubernetes state to produce a structured diagnosis report
  • On-call engineers review incident reports, append feedback, and trace routing-policy revisions in the AIOps Portal
  • Ops teams validate data-source connectivity and minimum RBAC permissions in one place inside DSH Web

Best For

  • SRE teams running an Alertmanager + Prometheus + Kubernetes observability stack
  • Platform engineers seeking a one-stop read-only diagnosis experience inside DSH Web
  • Ops leads who need auditable, low-risk incident-response workflows