Rolloutly: a deploy dashboard for a platform team running a dozen services
Deploy pipelines with streaming logs, production rollbacks, live service health, and charts of deploys per day, build time, success rate and time to recovery.
Agent specs
- Agent
- Claude Code, Opus 5.5 Medium
- Time
- 29 min
- Cost
- $9.68*
Agents work faster and more accurately on Elements.
* Cost at API rates, September 2026.
Run it yourself
Paste this prompt into your AI agent. You get the finished app on your own machine, ready to change however you like.
Install elements: curl -fsSL 'https://install.elements.dev?os=darwin' | sh && export PATH=~/elements/bin:$PATH Create an Elements app from the elementscode/demo-rolloutly scaffold: cd ~/elements/projects elements create rolloutly -scaffold=elementscode/demo-rolloutly cd rolloutly Start the app.
How it's built
Rolloutly needed deploy pipelines that stream their logs, production rollbacks, service health that refreshes itself, charts of a month of deploys and roles for the team. Each of those is a part of Elements, so the agent spent its 29 minutes on the dashboard itself.
What Elements gave the app
- Streaming deploys. A deploy is a job that walks each stage and sends every log line, stage change and final status on a channel. The deploy page prints the log as it runs, and the overview, service and activity pages pick up each status change.
- Live service health. A one-line cron schedule runs a health job every minute that takes a reading every ten seconds and pushes it to every open screen on a second channel. A release that failed its health check stays degraded until a new deploy or a rollback replaces it.
- Rollbacks. Rolling back checks that the chosen release ran in production and that nothing else is deploying, then sends it through the same pipeline as a deploy.
- Charts from the data. The insights page computes deploys per day, build time, success rate and time to recovery in SQL and draws them as SVG, and refreshes when a deploy finishes.
- Server calls as function calls. Starting a deploy, rolling back and managing the team call server functions straight from the page with
@rpc. - Data and roles from SQL. Migrations define the platform and seed one admin, three engineers, twelve services and thirty days of deploy history with failures and rollbacks. Sessions and roles give admins the team page.
What the project server gave the agent
The project server runs alongside the agent and answers as soon as a file is saved: it type-checks the templates, TypeScript and SQL, applies migrations and reruns the tests, so every question came back right away and the agent kept building.
What shipped
The app type-checks with zero errors and all 30 tests pass. Every page works on desktop and phone.
The code
Build it from scratch
To watch your agent build this app from an empty project, use the prompt we used instead. It takes longer than starting from the finished app above.
Install elements: curl -fsSL 'https://install.elements.dev?os=darwin' | sh && export PATH=~/elements/bin:$PATH Create an Elements app: cd ~/elements/projects elements create rolloutly cd rolloutly Start the app, then build this: Build a deploy dashboard named rolloutly for a platform team running a dozen services. Accounts: admin and engineer. - Services, each with environments (staging, production), the release running in each, and its health. - Deploy: pick a service, environment and commit. A background job runs the pipeline stages (build, test, migrate, release, health check), each taking a few seconds, with log lines streaming as it goes. Some deploys fail at a random stage so the dashboard has failures to show. - Deploy detail: stages with durations and the full log. - Roll back production to any earlier release, with a confirm step. - Charts, drawn in SVG: deploys per day, build duration over time, success rate, and mean time to recovery after a failed production deploy. - An activity feed of every deploy and rollback across all services. Seed one admin, three engineers, twelve services, and thirty days of deploy history with some failures and rollbacks. Show the seeded logins on the sign-in page. Running deploys, log lines, health and charts update in real time on every open screen.
Comments· 0