levelrail
Self-hosted alternative to Vercel and Coolify. Git push, HTTPS, logs, rollback. No Kubernetes.
Automatically checked · GitHub checked 2026-11-12
Open-source alternative to

Overview
Levelrail
Lightweight, self-hosted deployment platform: git push to production on your own servers and cloud accounts.
Git deploys, 570+ templates in the catalog (346 proven on a real engine), edge and TLS, nodes at six clouds, managed databases with restore-drilled backups, and built-in observability. Lightweight: the control plane idles at about 200 MB of RAM and 1.5 percent of one CPU on a 2 vCPU box, one Go binary ([measured idle footprint](docs/performance.md), [method](docs/benchmarks.md)). Apps you deploy use their own resources on top. Apache 2.0.
Docs · Features · FAQ · Discussions · Discord
Website · Docs · Install · Compare · FAQ · Roadmap · Discussions
English · 简体中文 · Español · Português (BR) · Deutsch · Français · 日本語

~200 MB control plane idle · 1 binary · 570+ templates in the catalog · 6 clouds · 13 database engines · 18 alert channels
Levelrail turns Linux servers into a private cloud and works with the Kubernetes clusters and cloud accounts you already have. Single-server is production ready, multi-node is beta. Every feature carries a stable, beta or experimental label in Feature status, and the labels below follow it.

What you get
Maturity follows Feature status. Open the list under each area for every capability.
- Push to GitHub, GitLab, Bitbucket, Gitea or Forgejo and it deploys
- Rollouts wait for readiness, rollback is one step
- Preview environments for every pull request
- Canaries and split tests before full traffic
All deploy features
- Git push or webhook deploys from GitHub, GitLab, Bitbucket, Gitea and Forgejo. See Git integrations and Git push deploy.
- Dockerfile, Compose and Railpack builds (Railpack is beta) with a build queue. See Build Center.
- Readiness-gated rollouts and one-step rollback. See Zero-downtime deploys and Rollback.
- Canaries and split tests. See Canary deploys and Split testing.
- Preview environments per pull request: stable on GitHub, beta on GitLab and Bitbucket. See Deploy previews.
- Deploy approvals and deploy freezes. See Deploy safety.
- Functions in Node, Python or Go (beta). See Functions.
- 570+ templates in the catalog, 346 with a recorded real-engine pass
- 22 categories, from databases to AI tools
- How templates work
All template details
- 574 catalog entries (565 active, 9 deprecated) in 22 categories.
- 346 entries have a recorded real-engine pass.
- 228 entries are marked preview: installable, with no recorded pass yet.
- Certificates issued and renewed for you, wildcards included
- Firewall, bot lists and attack mode built in
- Headers, redirects, geo rules and caching per domain
- Private networks between nodes (beta)
All edge features
- Embedded ingress with automatic TLS through HTTP-01 and DNS-01 wildcards.
- Per-domain headers, forwarders, geo rules, caching and redirects. See Domain traffic controls.
- Firewall rules, IP lists, browser challenge, bot lists and attack mode. See Firewall.
- Per-app rate limits and deny lists (beta). See Abuse defense.
- Round-robin balancing across replicas. See Load balancing.
- Private networks and a WireGuard mesh (beta). See Multi-node.
- Provision nodes at 6 clouds (beta)
- Join any Linux host with a single-use token
- DNS and registries at the clouds you already use
- Existing Kubernetes clusters as deploy targets (experimental)
All node and cloud features
- Provision nodes at Hetzner, DigitalOcean, AWS, Azure, GCP or Oracle Cloud (beta).
- Join any Linux host with a single-use token over mTLS. See Add a node anywhere.
- DNS at Cloudflare and Route53, image registries on ECR, Artifact Registry and ACR. See Cloud integrations for per-cloud status.
- Works with the clusters and cloud accounts you already have (experimental): connect existing Kubernetes clusters and deploy to them, off by default. See Kubernetes clusters.
- 13 database engines from one place
- Backups to S3-compatible, R2 or local disk
- Restore drills prove a backup works
- Point-in-time recovery for Postgres
All data and backup features
- 13 database engines in code. 9 are live-verified. 4 (Meilisearch, Qdrant, Typesense, Memcached) are built but not yet live-verified.
- Backups to S3-compatible, R2 or local targets. See Backups and storage.
- Restore drills that prove a backup works. See Prove your backups work.
- Point-in-time recovery for Postgres.
Built-in metrics, logs and alerts
- Alerts to 18 channel kinds, from Slack to PagerDuty
- Metrics and log search with no extra stack
- OTLP ingest for your own telemetry
- Public status page for your users
All observability features
- Built-in metrics, log search and log archive. See Log explorer and Log archive.
- OTLP ingest. See OTLP ingest.
- Public status page. See Status page.
- Deploy traces and traffic export. See Deploy trace and Traffic export.
- Alerts to 18 notification channel kinds. See Alerts.
- HTTP, TCP, heartbeat and TLS uptime monitors (experimental). See Monitors.
- Sign in with 20 OAuth providers plus generic OIDC
- Passkeys and TOTP for local accounts
- IAM policies and teams
- Hash-chained audit log you can verify
All security and access features
- Passwords, passkeys and TOTP.
- OAuth sign-in with 20 providers plus generic OIDC. See OIDC.
- SAML and SCIM (beta, off by default). See SAML and SCIM.
- IAM policies and teams. See Teams and roles.
- Hash-chained audit log. See Audit log.
- Encrypted secrets. See Environment and secrets.
- Signed releases. See Verify a release.
- 259 MCP tools for coding agents
- Terraform and OpenTofu provider, on the Terraform registry and the OpenTofu registry
- HTTP API and
levelrail-cli - 11 per-app integrations, from error tracking to analytics
All automation features
levelrail-cliand an HTTP API. See API reference.- Terraform and OpenTofu provider. See Terraform provider, the Terraform registry and the OpenTofu registry.
- YAML platform as code (experimental). See Platform as code.
- CI pipelines. See Pipelines.
- 11 per-app integrations: Sentry, PostHog, GlitchTip, OneUptime, Plausible, Datadog, Axiom, Better Stack, LogSnag, Bugsnag and New Relic. See Alerts.
- Signed plugins for notification channels. See Plugins.
- 259 MCP tools for coding agents. See MCP and Tool surface.
- A mail server template with mailboxes and DNS records (experimental). See Mail template.
All features, with maturity labels
Runs on Linux amd64 and arm64 with systemd and Docker (benchmarks).
Install
curl -fsSL https://levelrail.com/install.sh | sudo sh
Needs Linux (amd64 or arm64) with systemd and root; the installer adds Docker. Open ports 80 and 443, open the setup link it prints, choose a password, then deploy from the CLI (the dashboard's Connect your CLI card gives you the login line; pass --api-url always, and --profile NAME if you manage more than one server):
levelrail-cli login --api-url http://127.0.0.1:8080 --code ABCD-EF23
levelrail-cli import https://github.com/your-org/your-app --deploy
![]() Install, then deploy |
![]() Gated deploy, one-step rollback |
First 15 minutes · Installing (Docker, pinned version, source build, troubleshooting) · CLI only (brew install glincker/tap/levelrail-cli or npx levelrail-cli) · Free Arm VM · Migrate from Coolify, Dokploy, CapRover or Dokku · Verify signed releases
Using a coding agent? npx skills add https://github.com/glincker/levelrail --skill levelrail, then ask it to deploy your repo. See Agents.
Install with your coding agent
Paste this into Claude Code, Cursor or any agent that can run shell commands on your server:
Install Levelrail on this server: run `curl -fsSL https://levelrail.com/install.sh | sudo sh`.
Open the setup link it prints and tell me, because I choose the password myself.
Then run `levelrail-cli doctor` and `levelrail-cli status` and report anything that fails.
Do not write any token or password to a file or into this chat.
All screenshots are captured from a real running control plane. More in the Product tour.
Get Levelrail everywhere
| Channel | What it is | Install or use |
|---|---|---|
| Signed binaries and release notes | curl -fsSL https://levelrail.com/install.sh | sudo sh | |
| Control plane image | docker pull levelrail/levelrail:<tag> | |
| Control plane and node agent images | docker pull ghcr.io/glincker/levelrail:beta | |
| The CLI, no server | npx levelrail-cli --help | |
| The CLI on macOS and Linux | brew install glincker/tap/levelrail-cli | |
Provider glincker/levelrail | source = "glincker/levelrail" then terraform init | |
| The same provider | source = "glincker/levelrail" then tofu init | |
| MCP server listing | AI assistant setup | |
| Go kit packages | go get github.com/GLINCKER/levelrail/kit |
All packages and registries
| Package or channel | What it is | Install or use | Status |
|---|---|---|---|
| levelrail-cli | The CLI on npm | npx levelrail-cli --help | Live |
| levelrail-cli-linux-amd64, linux-arm64, darwin-amd64, darwin-arm64, windows-amd64, windows-arm64 | The six platform binaries that levelrail-cli pulls in | Installed for you by levelrail-cli | Live |
| levelrail-mcp | The MCP server on npm | npx -y levelrail-mcp (works from the next release) | Name reserved |
| levelrail-mcp-linux-amd64, linux-arm64, darwin-amd64, darwin-arm64, windows-amd64, windows-arm64 | The six platform binaries that levelrail-mcp will pull in | Installed for you by levelrail-mcp | Name reserved |
| ghcr.io/glincker/levelrail | Control plane image | docker pull ghcr.io/glincker/levelrail:beta | Live |
| ghcr.io/glincker/levelrail-agent | Node agent image | docker pull ghcr.io/glincker/levelrail-agent:beta | Live |
| Docker Hub levelrail/levelrail | Mirror of the control plane image, it lags GHCR (latest tag today: v1.3.0) | docker pull levelrail/levelrail:v1.3.0 | Live |
| Homebrew tap | The CLI on macOS and Linux | brew install glincker/tap/levelrail-cli | Live |
| Scoop bucket | The CLI on Windows, the manifest arrives with the next release | scoop bucket add glincker https://github.com/glincker/scoop-bucket then scoop install levelrail-cli | From the next release |
| winget | The CLI on Windows, not yet submitted to winget-pkgs | winget install Glincker.LevelrailCli (does not install today) | Planned |
| levelrail-agent .deb and .rpm | Node agent packages for Debian, Ubuntu, Fedora and RHEL, amd64 and arm64 | Download levelrail-agent-linux-<arch>.deb or .rpm from the release page | Live |
| levelrail-cli .deb and .rpm | CLI packages attached to each release | sudo dpkg -i levelrail-cli-linux-amd64.deb or sudo rpm -i levelrail-cli-linux-amd64.rpm | From the next release |
| AUR and Nix files | levelrail-cli-bin.PKGBUILD and levelrail-cli.flake.nix, attached to the release | Download from the release page | From the next release |
| Terraform Registry | Provider glincker/levelrail | source = "glincker/levelrail" then terraform init | Live |
| OpenTofu Registry | The same provider | source = "glincker/levelrail" then tofu init | Live |
| pkg.go.dev | Go kit packages | go get github.com/GLINCKER/levelrail/kit | Live |
| Glama | MCP server listing | AI assistant setup | Live |
| Official MCP Registry | Entry io.github.glincker/levelrail | AI assistant setup | From the next release |
| GitHub Releases | Signed binaries and release notes | curl -fsSL https://levelrail.com/install.sh | sudo sh | Live |
| install.sh | Installer script for the control plane | curl -fsSL https://levelrail.com/install.sh | sudo sh | Live |
Who it is for
Whoever you are, you start with one line.
| You are | You get | Start |
|---|---|---|
| A solo developer with one VPS | Git deploys, HTTPS in one click and about 200 MB idle | curl -fsSL https://levelrail.com/install.sh | sudo sh |
| Building with Claude Code, Cursor or another coding agent | An MCP server that deploys, waits, diagnoses and rolls back inside the token's limits | npx skills add https://github.com/glincker/levelrail --skill levelrail |
| An agency or team | Approvals, deploy freezes, IAM policies, a verifiable audit log, OIDC sign-in | Identity and access |
| On Coolify, Dokploy, CapRover or Dokku today | A dry run first, and the old server keeps serving | levelrail-cli import platform coolify --url https://coolify.example.com --dry-run |
| Running a homelab or a free Arm VM | arm64 builds and a minimal profile | Server profiles, Arm VM |
Built for coding agents
Let your agent ship. You keep the keys.
Deploy this repo to my Levelrail and tell me when it is live.
- Inside the token's limits. The tool list follows the abilities of the token: a read token never sees a deploy tool, and
--readonlyrefuses to start with a write token. - Writes need a confirmation. Mutations are confirm-gated and off until you pass
--allow-write. - Cheap to load. About 2.4k tokens of tools by default, 259 tools in total, on demand in toolsets (tool surface).
- Audited. Every call is recorded with the caller surface (CLI, dashboard, MCP or API).
- Not in the control loop. AI never sits in the reconcile path, so a wrong guess cannot wedge your servers.
- A skill, not just tools. It deploys, waits for the rollout to converge, reads the failure and rolls back (skill, MCP, threat model).
vs Coolify, Dokploy, CapRover and Dokku
Same job, fewer moving parts, more proof.
| Levelrail | Coolify | Dokploy | CapRover | Dokku | |
|---|---|---|---|---|---|
| Agent access (MCP) | Read and write, ability-scoped, confirm-gated, audited | Official server, read-only | MCP server repo exists | Not documented | Not documented |
| Switching in | Dry-run importers for all four, old server keeps serving | n/a | n/a | n/a | n/a |
| Node control | Reverse-dialed agent | SSH + CLI | SSH + Engine API | Swarm API | Local bash |
| Built-in metrics and logs | Yes | Opt-in add-on | Separate binary | No | No |
| Rollback | Pinned images, readiness-gated | Retained images | Swarm rollback | Manual | Default scheduler: none |
| Backup verification | Automatic | Non-empty check | None | None | No backups |
| Audit log | Hash-chained, verifiable | Not documented | Not documented | Not documented | Not documented |
Cells state only what each project's public docs say (last checked 2026-11-07), and "Not documented" means we found nothing, not that it is missing. Coolify's MCP: their docs. This is positioning, not a ranking. Levelrail is the youngest project here, with fewer users and less production mileage. Sourced details: Comparison.
Production ready, with receipts
| Claim | Evidence |
|---|---|
| Every feature labelled stable or beta | Feature status |
| Survives failures | Resilience: disk full, link loss, database damage, safe-mode boot |
| Measured, not promised | Benchmarks |
| Tested against real containers | CI and test/e2e |
| Signed releases | Sigstore signature and provenance attestation, verify them |
| Honest about gaps | Known limits |
How it works
flowchart LR
C["Dashboard, CLI, MCP server"] --> API
subgraph CP["Control plane (one Go binary)"]
API["HTTP API"] --> R["Level-triggered reconciler"]
R --> S[("SQLite state")]
O["Built-in metrics and logs"]
I["Embedded Caddy ingress and TLS"]
end
R -->|"gRPC, mTLS"| A["Node agents (dial out)"]
A --> D["Docker Engine API"]
I --> D
Details: Architecture.
Guides
Next.js with Postgres · Django with Postgres · Push to deploy with GitHub · How deploys stay safe · Scale to zero on a small VPS
Maintain an open-source app? Add a one-click button: Deploy button.
FAQ
Does it need Kubernetes? No. The control plane is one Go binary with a Docker runtime. If you already run Kubernetes, you can connect a cluster and deploy to it (experimental, off by default). See Kubernetes clusters.
What does it cost? Nothing. Apache 2.0, one edition, no license key. You pay for your own servers (Pricing).
Can I migrate from Coolify? Yes. levelrail-cli import platform coolify --url https://coolify.example.com --dry-run (with the source token in APP_IMPORT_SOURCE_TOKEN) shows what would be created first, and the old server keeps serving (Migrate).
How do I upgrade? Settings > Updates or levelrail-cli upgrade --plan. It verifies the release, snapshots your data and rolls back by itself if the new release is unhealthy (Upgrade safely).
Community
Discussions · Issues · Discord · [email protected] · Security policy · Support · Accessibility · Versioning and version tracker · Contributing · llms.txt and AGENTS.md for AI agents
Star the repo
If Levelrail saves you a weekend, a star helps other self-hosters find it.
Apache 2.0, see LICENSE. Built by Gagan Deep Singh of GLINR STUDIOS and theSVG.
Powered by GLINR STUDIOS · icon from theSVG







