WireZTNA is now open source
Today we're releasing WireZTNA under Apache 2.0: the full Zero Trust platform, built on WireGuard, self-hosted, with every feature. No stripped-down edition, no artificial limits, no lock-in.
What we released
WireZTNA is a Zero Trust Network Access platform built on WireGuard: it replaces the flat VPN with fine-grained, per-user access to private resources, following the model User → Group → Publisher → exposed resources. As of today, the code is publicly available under the Apache 2.0 license on GitHub.
This is not a watered-down "community" edition. It's the complete platform, with every feature:
- Broker — the control plane and overlay relay (WireGuard, nftables enforcement, per-publisher isolation with Linux network namespaces).
- Publisher — the agent you drop into each private network (Go binary or Docker). Connects outbound; no inbound ports to open.
- Multi-platform clients — Linux, macOS, Windows (CLI + TUI), a desktop app, and Android. All source code included.
- wzctl — the CLI for ephemeral, scoped access to a single
host:port(TTL + revocation + audit). Ideal for CI and automation. - Admin web UI — manage users, groups, publishers, and access.
The access model, enforcement, and data path are identical to what WireZTNA runs in production. The open source edition differs only in packaging and defaults (bare public IP, HTTP, username + password login), not in capability.
Why Apache 2.0
We chose Apache 2.0 for adoption and trust. It's a permissive license with an explicit patent grant, which makes it suitable for companies that want to integrate it without legal friction. We want anyone — from a homelab to a platform team — to be able to deploy WireZTNA on their own infrastructure, audit it, modify it, and depend on it with no strings attached.
- No license key. Nothing to activate.
- No usage limits. Self-hosted means it's yours: as many users, publishers, and tunnels as you want.
- No phone-home. Your data and your traffic never leave your network.
Built to deploy in minutes
The self-hosted edition is designed to run on a single Linux host with just a public IP — no domain, no TLS certificate, no cloud account. The installer builds nothing on the broker: you compile the artifacts on a dev box and the broker serves them.
# On your broker host (a Linux VPS with a public IP)
sudo ./install.sh --ip <YOUR_PUBLIC_IP>
# Create the first admin and open the UI at http://<YOUR_PUBLIC_IP>/
# Deploy a publisher inside the private network you want to reach:
sudo ./wireztna-publisher install --token "http://<YOUR_PUBLIC_IP>/api/v1/publishers/enroll?token=<token>"
Username + password login is the default: you don't need to configure email or a validated domain to get started. Email OTP and SSO/OIDC remain available once you configure them.
So where's the business?
If everything is open source and self-hosted, what do we sell? Deliberately, we don't sell crippled features. The only paid advantage is aimed at enterprise environments that don't want to operate the platform themselves:
- Managed SaaS — we operate the broker, with managed availability, updates, and scaling. You deploy and maintain nothing.
- Premium support — SLAs, direct engineering support, and hands-on help for mission-critical deployments.
In other words: the technology belongs to everyone; what we charge for is running it for you and being there when you need it in production. A team that prefers to self-host has exactly the same capabilities, free, forever.
Get started
- ⭐ Star and clone the repository: github.com/wireztna/wireztna-oss
- 📖 Explore the code with DeepWiki: deepwiki.com/wireztna/wireztna-oss
- 💬 Join the community on Discord: discord.gg/vqC82g2G
If your organization needs the managed service or premium support, let's talk. For everything else, the code is yours.