>publichost

PublicHost Documentation

PublicHost turns your local development server into a public HTTPS URL with a single SSH command. No account, no installation, no configuration — just an SSH client that you already have.

  1. 1. Install the CLI

    One command, no accounts. See Installation below for Linux, macOS, and Windows.

    bash
    curl -fsSL https://publichost.run/install.sh | sh

  2. 2. Start something on localhost

    For a quick test run a local HTTP server on port 3000:

    bash
    # Python
python3 -m http.server 3000

# Node
npx serve -p 3000

  3. 3. Open a tunnel

    Replace 3000 with the port your application is running on:

    bash
    publichost run 3000

    Prefer raw SSH, or don't want to install anything? The CLI is a thin wrapper around the same connection:

    bash
    ssh -p 2200 -R 0:localhost:3000 publichost.run

  4. 4. Use the URL from the banner

    Once connected, PublicHost prints a banner similar to:

    bash
    ============================================================
Your tunnel is ready:
 https://swift-fox-123.publichost.run

Live logs:
 https://publichost.run/channel/weblogs?token=abc123
============================================================

    Open https://swift-fox-123.publichost.run in a browser or share it with a teammate.

  5. 5. Close the tunnel

    Press Ctrl+C or close the terminal. The subdomain is released automatically and will never be used again.

Installation

Requires Python 3.9+ and an ssh client, both already on most systems. The installer is idempotent — running it again upgrades the CLI in place.

Linux / macOS

bash
curl -fsSL https://publichost.run/install.sh | sh

Windows (PowerShell)

powershell
irm https://publichost.run/install.ps1 | iex

Windows 10/11 ships an OpenSSH client that can be enabled from Settings → Optional Features if ssh isn't already on your PATH.

Run it

Once installed, open a tunnel to any local port — same command on every OS:

bash
publichost run 3000
publichost run 8080 --subdomain my-app

CLI Usage

Basic tunnel

bash
ssh -p <ssh_port> -R 0:localhost:<local_port> publichost.run
  • <ssh_port> — the SSH port of the PublicHost server (commonly 22 or 2200).
  • <local_port> — the port your local service is listening on.
  • publichost.run — the SSH domain of the server.

Example

bash
ssh -p 2200 -R 0:localhost:8080 publichost.run

No authentication required

PublicHost does not ask for a password or a key. The connection is accepted immediately and a random subdomain is assigned to you.

One-time subdomain

Every SSH connection receives a fresh, randomly generated subdomain such as swift-fox-123.publichost.run. The subdomain lives only as long as the SSH connection. When you disconnect, it is released and never reassigned.

Tunnel lifetime

By default a tunnel is kept open for up to 2 hours. If you need more time, extend it from the interactive banner session:

bash
extend 60

This adds 60 minutes to the current tunnel lifetime. You can extend multiple times while the connection is active.

Multiple local services

You can request several remote forwards in a single SSH connection:

bash
ssh -p 2200 \
 -R 0:localhost:3000 \
 -R 0:localhost:8080 \
 publichost.run

Each forward gets its own subdomain.

Running in the background

Use -N to open the tunnel without starting an interactive shell:

bash
ssh -p 2200 -N -R 0:localhost:3000 publichost.run

Add -o ServerAliveInterval=60 to keep the connection alive through NATs and firewalls:

bash
ssh -p 2200 -N -o ServerAliveInterval=60 -R 0:localhost:3000 publichost.run

Persistent tunnels with autossh

For long-running tunnels install autossh and run:

bash
autossh -M 0 -p 2200 -N -o ServerAliveInterval=60 -R 0:localhost:3000 publichost.run

Live Logs

PublicHost records every HTTP request that travels through your tunnel and streams it to a WebLogs page in real time.

Opening Live Logs

The tunnel banner contains a link like:

bash
https://publichost.run/channel/weblogs?token=abc123

Open that URL while the tunnel is active to see requests as they happen.

What is shown

Each row in the live log contains:

  • Timestamp — when the request was received.
  • Method — HTTP method (GET, POST, PUT, DELETE, etc.).
  • Path — the request path and query string.
  • Status — the HTTP status code returned by your local service.
  • Duration — how long the request took.
  • Client IP — the original remote address.

Scope

Live logs belong to a single tunnel session. When the SSH connection closes, the log stream stops and the token becomes invalid. A new tunnel gets a new token.

Downloading logs

Completed tunnel sessions are archived as CSV files on the server. If you need historical logs, contact the server administrator or use the backend API.

Troubleshooting

Connection refused on the SSH port

  • Check that you are using the correct SSH port.
  • Verify the server firewall allows inbound connections on that port.

502 Bad Gateway for the tunnel URL

  • Make sure your local service is still running on the port you forwarded.
  • Make sure the SSH connection is still active.

Subdomain does not resolve

  • DNS can take a few minutes to propagate.
  • Confirm that *.publichost.run resolves to the PublicHost server IP.

Security Notes

  • Subdomains are single-use and expire with the connection.
  • No shell, exec, SFTP, or other SSH features are exposed — the session is banner-only.
  • Do not share your live-logs token publicly; it grants read access to request metadata for that tunnel.