Free, open source

DTeploy. Observe. Build.
Own your stack.

Free tools to deploy, monitor, and develop. Zero-downtime deploys via SSH, observability, and a terminal IDE. No lock-in, no hosted dependencies.

Works with
VultrHetznerDigitalOcean
How it works

Four steps to production

01

Install

One command. Homebrew or shell script. macOS, Linux, WSL.

brew install useteploy/tap/teploy
02

Connect

Point at any server via SSH. No hosted dependencies.

teploy setup 203.0.113.10
03

Deploy

Zero-downtime deployment with instant rollback.

teploy deploy
04

Monitor

Uptime checks with Dash. Analytics, APM, and errors with Observe.

teploy-observe
Automatic HTTPSSelf-hosted observabilityNo vendor lock-in
Architecture

Your infrastructure, simplified

teploy — topology
ssh deployshttps trafficClick to deployTeploydeploys over sshServer 1vultr — nyc3 containersServer 2hetzner — eu5 containersServer 3do — sfo2 containersUsersserved over https — caddyhover nodes · click teploy to deploy
CLI

Deploy to any server, zero downtime

Single Go binary. SSH into any server. Three lines of config. The engine behind every Teploy install.

myapp — teploy

$ teploy deploy

Building from Dockerfile...

Pushing myapp:9f3c21 to registry

Starting new container on port 3001

Health check passed (200 OK, 42ms)

Routing traffic via Caddy (HTTPS auto)

Stopping old container

Deployed in 14s. Zero downtime.

Zero-downtime deploys

New container starts, health checks pass, traffic routes, old container stops. Automatic rollback on failure.

Caddy auto-HTTPS

TLS certificates provisioned and renewed automatically. Routing and load balancing configured for you.

Multi-server in parallel

Deploy to multiple servers at once with load-balancer configuration. Preview environments with auto-expiry.

Accessories & backups

Postgres, Redis, MySQL alongside your app. S3 backups with cron scheduling and database-aware dumps.

Dash

A web UI that never desyncs

Run it locally for a quick look, or standalone for 24/7 monitoring. Reads the same state files the CLI writes.

teploy-dash — fleet
4 apps / 3 servers
apihetzner-eu
running99.99%
webvultr-nyc
running99.98%
workerdo-sfo
deploying99.97%
postgreshetzner-eu
running100%

Live fleet view

Every app across every server — status, image hash, domain, container state. Deploys stream in real time over WebSocket.

Apps & accessories

Start, stop, restart, rollback. Manage databases and caches alongside your apps from one screen.

Uptime monitoring

Standalone mode adds HTTP/TCP health checks, response-time graphs, and an incident timeline.

Alerts on outages

Webhook, email, or Slack notifications when a check fails or recovers. Per-monitor thresholds.

Observe

See what's happening

Analytics, APM, error tracking, and session replay in one self-hosted binary. Privacy-first.

observe — myapp.com
Visitors 12.4kp95 42msErrors 0.02%
00:0006:0012:0018:00now

Web analytics

Page views, sessions, referrers, devices. Lightweight script, no cookie banners needed. Everything connects — click an error, see the trace.

APM & tracing

OpenTelemetry-native distributed tracing. Latency percentiles, slow query detection, service maps.

Error tracking

Catch exceptions before users report them. Stack traces, source maps, affected-user counts.

Unified dashboard

Click an error, see the trace. Click a slow page, see the API calls. Everything connects automatically.

Ship

Issues in, verified PRs out

A controlled agent workflow: automation where it is useful, review where it matters. Self-hosted.

ship — myapp

$ teploy ship fix "flaky test in api/auth"

plan read 12 files · found a race in token refresh

act edit api/auth/session.go (+8 -3)

obs go test ./... 47 passed

act commit · push branch ship/auth-race

Opened PR #128 → fix: guard token refresh race

Submit, work, verify, review

Ship takes a coding task, works it on a branch, and opens a pull request that carries proof it works.

Issue webhooks

Label an issue in Forgejo or GitHub and a worker picks it up - clone, branch, and get to work.

Verified before review

Tests and checks run after the agent stops writing. The PR carries the evidence, not promises.

Caps and approvals

Spend limits, plan preview, and review gates. The agent works; you stay in charge.

3

lines of config to deploy

14s

to a zero-downtime deploy

1

binary per tool, no dependencies

4

steps from issue to a verified Ship PR

Free. Open source. Yours.

Deploy, monitor, and develop with tools you own. No lock-in, no subscriptions required.

$brew install useteploy/tap/teploy

CLI + Dash + Observe free forever. Ship brings the agents.