Deployment Dashboard

A real-time services × environments deployment matrix, sourced straight from your CI/CD pipeline events.

Get started Integrate your CI/CD GitHub Built by Claude — see how

MatrixSwimlanesFeedAnalyticsServices filterNotificationsProvided presetsInstall as an app

One row per service, one column per environment. Each tile shows version, status, actor, elapsed time, and a link to the CI/CD run.

Per-service deployment graphs — see how a version flows from dev through to prod, with branching topology and status-colored edges.

A chronological log of every deployment event, grouped by deployment with expandable roll-ups, server-side search across the full history, and cursor-based infinite scroll. A toggleable bottom dock mirrors the same log — last 8 events, live via SSE — on every other view.

DORA Four Keys — deployment frequency, lead time, change failure rate, and MTTR — plus eight supporting charts to spot delivery trends and regressions over 7, 14, or 30 days.

Narrow the board to exactly the services you care about — type a glob pattern (front-*, api, or a namespaced org-a/gateway), choose "Show only" or "Show all except", and the Matrix rows and Swimlanes lanes update instantly. Patterns persist across reloads. When the same service name exists under different namespaces (e.g. org-a/gateway and org-b/gateway), each appears as a distinct row with its namespace/ prefix shown only when there is a collision. A slashless pattern matches across all namespaces — existing saved patterns keep working without change. The same widget powers the notification service and environment filters.

Opt-in desktop notifications for every status transition — filtered by status, service, and environment. Enable the bell toggle once; the browser stays in the loop even when the tab is in the background.

Repo/CI-sourced UI-settings presets, published by any owner/repo source — over a plain curl PUT or auto-discovered by the optional Fetcher — and shown read-only to every visitor in a PROVIDED section of the presets popover, with one-click apply.

Provided presets guide

Install the dashboard as a standalone app in Chrome or Edge — dedicated window, taskbar icon, no browser chrome. Works behind your gateway; no extra deployment needed.

How to install

The question it answers: What version of service X is running in environment Y right now — and did the last deployment succeed?

Why Deployment Dashboard?

• No pipeline rewrite

  ---

  Integration is a single HTTP POST step. No plugins, no agent to install, no migration — drop one step into the pipeline you already have.

• Tool-agnostic

  ---

  GitHub Actions, Azure DevOps, Jenkins, GitLab CI — or a shell script. If it can call a URL, it can feed the dashboard. The backend never knows which tool you use.

• Answers one question, instantly

  ---

  What's running in prod right now, and did it succeed? One screen — every service across every environment, no clicking through pipelines and logs.

• Live, not polled

  ---

  Server-Sent Events push every state change to every open browser within seconds. The matrix updates itself — no refresh, no stale tab.

• Stay notified — without watching the screen

  ---

  Enable the bell toggle once and get a desktop notification on every deployment status change, even when the tab is backgrounded. Filter by status, service, and environment. No browser extension required.

• Full history, append-only

  ---

  Every deployment is kept per slot (≥ 90 days, configurable). Nothing is overwritten; the history drawer shows the entire timeline of a slot.

• Auto-discovers your topology

  ---

  Services and environments are derived from the events you send. No registration, no config file, no hardcoded lists — post a new service and it appears.

• Stateless & cheap to run

  ---

  Scale API instances behind the gateway with no sticky sessions. Runs on any OCI container host; the reference Azure target fits in ~$30/month.

• Secure by design

  ---

  Writes are API-key gated; reads are internal-only and the SPA holds no secrets. You own the network boundary — nothing is public by default.

• Pull mode when you can't push

  ---

  Can't touch the pipeline? The optional Fetcher polls your CI/CD API and posts on your behalf — through the very same contract.

  Pull mode also fits locked-down networks that forbid inbound WAN traffic: the Fetcher is outbound-only (it calls the GitHub API and the dashboard's internal ingest), so the dashboard never has to accept inbound connections from the internet — unlike push, where CI/CD must reach in to POST.

Send your first deployment

One HTTP call from your pipeline — that's the whole integration:

curl -X POST "$DASHBOARD_URL/api/deployments" \
  -H "X-Api-Key: $DASHBOARD_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "deployment_id": "build-42",
    "service":       "checkout",
    "namespace":     "storefront",
    "environment":   "prod",
    "version":       "1.4.2",
    "status":        "success",
    "happened_at":   "2026-06-01T10:00:00Z"
  }'

GitHub Actions, Azure DevOps, GitLab & Jenkins examples

Architecture at a glance

A handful of containers behind one gateway: a stateless .NET API tier, PostgreSQL as the append-only source of truth, and an optional pull-mode Fetcher. The gateway is the only public surface — reads are internal-only and the SPA holds no secrets.

Architecture overview & the security model

Explore the docs

•   Quickstart

  Run the whole stack locally in two minutes, zero config.

•   Install & deploy

  Compose profiles, production checklist, and Azure/Terraform Container Apps deploy path.

•   Configuration

  Every environment variable, grouped by concern.

•   Architecture

  How the pieces fit, and the security model.

•   API contract

  OpenAPI spec + human-readable guidelines.

•   Contributing

  Local setup, branch → PR workflow, conventions.

---

Looking for the full specification? The Development & reference tab covers the architecture (SAD), API contract, frontend design, services, and diagrams.