The parts you need
to run what you build.

Add a service, a database or a domain to a project and it runs, already wired to the rest. You only pay for what it uses.

Builds

From a repository or an image to something that runs, no Dockerfile required.

  • Automatic buildsLucity detects the framework and builds it for you.
  • Prebuilt imagesAny public image, like Ghost or Grafana, found with a search in the dashboard.
  • Build cacheEvery build starts from the last one, so only what changed is rebuilt.
  • MonoreposPoint each service at its own folder of the same repository.
  • Build settingsOverride the install, build and start commands, or add system packages.
  • Framework guidesNext.js, Nuxt, SvelteKit, Astro, Express, FastAPI, Django, Laravel and Go.

Deploys

Getting a new version live, and back again if it misbehaves.

  • Push to deployEvery push to the tracked branch deploys, and one commit can roll out every service built from that repository.
  • Deploys from GitHub ActionsDeploy once your tests pass, with no token stored anywhere.
  • Zero-downtime rolloutsThe new version takes traffic once it's healthy, and the old one finishes its requests.
  • Changes in secondsVariables, resources and the start command apply without a rebuild.
  • Failure reasonsA failed rollout says why, like out of memory or a crash loop, while the old version keeps serving.
  • RollbacksBack to a previous deployment when a release goes wrong.
  • EnvironmentsDevelopment, staging and production as separate copies of one project.

Services

How each part of your app runs: its size, its count and its settings.

  • Start commandKeep the detected one or set your own.
  • Health checksA connection check by default, or an HTTP check with extra time for slow starts.
  • CPU and memorySet each one separately instead of picking a fixed package, and change them without a rebuild.
    Hobby 2 vCPU, 4 GBPro 4 vCPU, 16 GB
  • ReplicasRun several instances, with traffic spread across the ones that are ready.
    Hobby 3 · Pro 10
  • AutoscalingMore instances as CPU climbs, fewer once it stays calm.
  • Private networkingServices reach each other by name, with no public address needed.
  • VariablesPer service, available while it builds and while it runs.
  • Shared variablesSet a value once for the whole environment.
  • ReferencesPick a database's connection string from a list instead of copying passwords around.

Domains

How visitors reach your app.

  • Platform domainsA public address with HTTPS for every service, ready in seconds.
  • Custom domainsYour own hostname with two DNS records, shown ready to copy.
    Hobby 1 · Pro unlimited
  • Apex domainsexample.com itself, not only www.
  • Automatic certificatesIssued as soon as a domain checks out, and renewed for you.
  • HTTPS onlyPlain HTTP requests are redirected to HTTPS.
  • Domain redirectsSend www to your apex, or an old domain to a new one, keeping the path.
  • Global edgeVisitors connect to a location near them, with nothing to change in DNS.

Postgres

Managed PostgreSQL with failover and backups on every plan.

  • High availabilityTwo instances on separate machines, with automatic failover.
  • Flexible sizingCPU, memory and storage are set one by one, not bundled into fixed small, medium and large packages. Change any of them without downtime.
    Hobby 32 GB · Pro 128 GB
  • Continuous backupsEvery write is archived as it happens and kept for 30 days, with nothing to configure.
  • Point-in-time recoveryRestore to any second of the last 30 days, into a new database next to the original.
  • Backups on demandTake a full backup before a migration you're unsure about.
  • Allowed addressesPrivate by default. Open it to the internet for the addresses you list, and nobody else.
  • Encrypted connectionsPublic connections are encrypted all the way to the database, not to a proxy in front of it.
  • Table browserRead tables and rows in the dashboard to check a migration did what you expected.
  • SQL editorRun queries against the database from the dashboard.
  • Database metricsConnections, transactions, cache hits and replica lag.
  • Top queriesThe statements taking the most time, so you know what to fix first.
  • Vector searchpgvector is included, for embeddings and similarity search.

Storage

Somewhere for everything that isn't rows in a table.

  • Key-value storeRedis-compatible, for caches, sessions and job queues.
    Hobby 32 GB · Pro 128 GB
  • BucketsS3-compatible storage for uploads and files.
  • Public bucketsServe files to everyone from a public address on a CDN.
  • File explorerBrowse, upload and download a bucket's files in the dashboard.
  • VolumesDisks that outlive deploys, for software that writes to its filesystem.
    Hobby 32 GB · Pro 128 GB
  • Volume resizingAdd space while the service keeps running.
  • Volume usageHow full a volume is, over time.

Logs and metrics

What your app is doing right now, and what it did last night.

  • Live logsFollow every instance as it prints, including the new ones a rollout starts.
  • Log historyLook back an hour, a day or a week, even for instances that are long gone.
  • Log searchFind a phrase across every instance, highlighted, with the lines around each match.
  • Instance filterNarrow the logs to one instance when several are running.
  • Log timelineHow much was printed over time, with errors in their own color, so spikes stand out.
  • Platform eventsStarts, restarts and failed health checks appear between your app's own lines.
  • Log downloadSave what you're looking at as a text file.
  • Deployment logsBuild and deploy output, streamed live and kept for a week.
  • Service metricsCPU and memory per service, over time.

Tools

Work from a terminal, a pipeline or your agent.

  • CLIDeploy, check status and manage variables from a terminal or any CI pipeline.
  • Claude plugin and MCPDeploy and debug from Claude Code, Codex, Cursor or any other MCP client.
  • APIEverything the dashboard does, with an interactive playground.
  • API tokensTokens for scripts and pipelines, each with its own role.

Team

Bring people in without sharing a password.

Billing

You pay for what your apps use, not for machines sitting idle.

  • Pay for what you useCPU and memory are billed on what your apps actually use, not a fixed price per app or machine, so an idle service costs very little.
  • Plan creditsYour plan's monthly fee comes back as usage credit, so you only pay for usage beyond it.
  • Usage so farWhat this billing period has cost so far, with an estimated total.
  • Free trial14 days to try everything, no credit card needed.

Security

On for every app, without asking.

  • Secret scanningEvery deploy checks your repository and its history, and flags keys that still work.
  • Vulnerability checksRunning apps Lucity built are rechecked every two hours.
  • Workspace isolationApps in one workspace can't reach another's apps or databases.
  • Environment isolationStaging can't reach production's private network.
  • Secret-free agent toolsAgent tools show variable names and references, never the values.

Ownership

Where it lives, and how you'd leave.

  • Hosting in the EUApps and databases in Germany, files and backups in France.
  • Open sourceRead every line of the platform.
  • Self-hostingRun the whole platform on your own infrastructure.
  • EjectDownload any project in a standard format and run it on any major cloud, without Lucity.

Coming from somewhere else?

The same list, side by side with what you use today.

Missing something you need? Tell us on GitHub or by email.

See what it costs