Documentation

EdgeVPN uses libp2p to build private, decentralized networks that are accessed with a shared secret (a token). There is no VPN server and no central coordinator: every peer that holds the token joins the same network and discovers the others over p2p.

A single statically compiled binary can:

• Create a VPN between peers. Each node takes its virtual address from --address; an experimental --dhcp mode lets peers negotiate addresses among themselves instead.
• Serve DNS for the network, if you enable it with --dns. It answers from the records peers announce on the shared ledger, and forwards anything it does not have to an upstream resolver.
• Act as a reverse proxy, exposing a TCP service to the network the way ngrok would, without bringing up a VPN interface.
• Send files directly between peers, again without a VPN interface.
• Be used as a Go library, so you can embed the same distributed ledger and p2p connectivity in your own program.

Where to go next

• Tutorials — start here if you are new. End-to-end walkthroughs that get you from nothing to a working network.
• How-to guides — task-oriented recipes: expose a service, proxy traffic through another peer, lock a network down.
• Reference — every command, flag, environment variable and API endpoint.
• Explanation — how EdgeVPN works and why, including the security model you should read before deploying it.

There is also a web UI and HTTP API for inspecting a running network (WebUI/API), and an alpha desktop GUI for Linux.