Multiple deployment sources
Deploy from GitHub, another Git host, a prebuilt Docker image, an uploaded archive, a Compose stack, or a catalog template. A three-step wizard fills in detected application details before deployment.
Deplo
Deplo is an open-source, self-hosted deployment platform for hosting web applications on servers you control. It provides dashboard-based deployments, automatic releases, previews, HTTPS, backups, logs, access controls, and optional AI operations without requiring routine Docker or SSH work.
Self-hosted deployments on your own infrastructure
Deplo is an open-source, self-hosted deployment platform for running web applications on servers you control. It provides a dashboard for deploying repositories, Docker images, archives, Compose stacks, and templates, while handling container deployment, routing, HTTPS, logs, metrics, backups, and server-agent communication.
The platform is designed to provide a push-to-deploy workflow without requiring users to manage Docker commands or SSH for routine tasks. It supports automatic deployments, pull-request previews, generated public URLs, framework detection, rollbacks, team permissions, and optional AI access through MCP. Deplo is currently in beta, and its documented installation targets Linux servers with Docker-compatible hardware.
Deployment and infrastructure features
Live framework detection
Framework, port, and build and start commands are derived from the repository on every deployment, so changes to the project setup do not require manually editing stored detection settings.
Automated releases and rollbacks
Deploy from a tracked branch or release tag, create deterministic per-pull-request preview URLs, and roll back by rerunning the exact image produced by a working deployment.
Self-hosted application platform
A single installation sets up Docker, the reverse proxy, Postgres, the control plane, and a server agent. Applications receive generated public hostnames, HTTPS, and separate environment networks.
Team access control
Teams can use server-side capabilities for individual actions, including separating deployment from deletion, and review an activity trail showing who acted and when.
AI-assisted operations
The native MCP server allows an authorized AI client to deploy, inspect logs, and roll back under a revocable token and the same permission checks as the dashboard. It is off by default.
Practical ways to use Deplo
Host a web application
Connect a repository, let Deplo detect the framework and commands, and publish the application with a generated public hostname and HTTPS without writing a Dockerfile or managing routine SSH steps.
Run a push-to-deploy workflow
Configure a tracked branch or release tag so pushes or new tags trigger deployments automatically. Use pull-request-specific preview URLs for review and remove them automatically when the pull request closes.
Migrate from another control panel
Move applications from another self-hosted panel using the migration wizard. The documented migration flow can recreate apps, databases, volumes, variables, domains, basic authentication, cron jobs, and backup schedules, with a report identifying apps that need review.
Delegate infrastructure tasks to an AI agent
Give a coding agent a revocable MCP token so it can deploy, read logs, or roll back while remaining subject to the permissions configured for the account or team.
Deploy packaged services and Compose stacks
Use templates or a Compose file to launch packaged services, with routing, HTTPS, variables, storage, and backups added around the application stack.
Pros and Cons
Pros
- Open-source, self-hosted deployment with applications and data hosted on infrastructure you control.
- Supports several deployment inputs, including repositories, Docker images, archives, Compose stacks, and tested templates.
- Built-in deployment automation, preview environments, HTTPS, backups, logs, metrics, and rollback workflows.
- Fine-grained team capabilities and an activity trail provide operational access control.
- Optional MCP support lets authorized AI clients perform deployment and troubleshooting actions under existing permissions.
Cons
- The product is currently in beta, with a stable release targeted for Q4 2026.
- Installation is documented for Linux servers and requires control of server networking, including internet-reachable ports 80 and 443.
- Some advanced workflows still have expert options such as a terminal and Compose escape hatch, so the simplest path does not cover every deployment configuration.
FAQ
How is Deplo installed, and what does a server need?
Deplo installs from a single shell command on a fresh Linux server. Supported documented options include Ubuntu 24.04 LTS or newer, Debian 12 or 13, and other distributions compatible with Docker; systemd is required for the server agent. The documented minimum is 2 CPU cores, 4 GB RAM, and 30 GB disk, with 80 and 443 reachable from the internet and 9443 available for agent hosts.
How do deployments work?
Connect a repository, choose a server, and use Deplo's deployment wizard. Deplo detects the framework, port, and build and start commands from the repository, then builds and serves the application. Deployments can also come from a prebuilt Docker image, uploaded archive, Compose stack, or template.
Does Deplo support automatic deployments and pull-request previews?
Yes. Deplo can deploy on pushes to a tracked branch or when a new release tag appears. It registers the webhook and shows whether the provider is reaching it. Open pull requests can receive their own preview URL, which is removed when the pull request closes.
Can an AI agent operate Deplo?
Deplo includes a native MCP server that can let Claude, Cursor, or another MCP client deploy, read logs, and roll back. MCP access is off by default, uses a token that can be minted and revoked, and follows the same permission checks as the dashboard.
Is Deplo production-ready and what does it cost?
Deplo is currently in beta, with a stable release targeted for Q4 2026 according to the site. It is open source under AGPLv3, and the site states that users pay their server provider rather than Deplo, with no per-seat pricing, build-minute charges, or bandwidth tiers described.
Quick Facts
- Category
- Self-hosted deployment platform
- Deployment model
- Open source and self-hosted
- License
- AGPLv3
- Documented host
- Linux server with Docker-compatible x86-64 or arm64 hardware
- Installation
- Single shell command on a fresh server
- Release status
- Beta; stable release targeted for Q4 2026
Альтернативы Deplo
ByteAsk
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
CreateOS Sandbox
CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.
GitBot
GitBot is a local browser-based workspace for creating reusable coding bots powered by Claude Code, Codex, or OpenCode. It helps developers run those agents across repository folders with persistent threads and configurable permissions.
hob
hob is an independent workspace for coding agents that keeps agent sessions, terminals, history, and follow-up work organized around the tools and providers you already use. It is aimed at developers who want local control over routing, history, and workspace structure rather than a bundled model stack.
Ably Chat
Ably Chat is a chat API platform for building custom realtime chat applications. It supports room-based messaging, typing indicators, presence, reactions, and message updates, with usage-based pricing options for different deployment stages.
Manta AI
Manta AI is an autonomous web app testing tool for teams that want to map application behavior, catch regressions, and generate tests without writing scripts or maintaining selectors. It works from a URL and supports plain-English test flows, run results with screenshots, and scheduled or deployment-triggered checks.