Setting up SSH on Windows without duplicating your work

Most Windows machines end up with two SSH clients: the one Microsoft ships and the one you installed. That is fine, as long as each has a clear job and you know which agent holds your keys.

The built-in OpenSSH client

It is an optional feature that is enabled by default on current builds. Its per-user config lives at %USERPROFILE%\.ssh\config and uses the same syntax as on any other platform:

Host jump
  HostName jump.example.com
  User admin
  IdentityFile ~/.ssh/id_ed25519

After that, ssh jump is all you type.

Turn on the agent service

Set-Service ssh-agent -StartupType Automatic
Start-Service ssh-agent
ssh-add $env:USERPROFILE\.ssh\id_ed25519

The service stores the unlocked key for the Windows session, which is the OpenSSH equivalent of Pageant.

Keep PuTTY and OpenSSH in their lanes

PuTTY for saved GUI profiles, serial consoles and Telnet devices.
OpenSSH for scripting, Git and anything expecting a config file.
Convert keys once in PuTTYgen so the same identity works in both.
Run one agent at a time to avoid confusing failures.

Client settings worth standardizing

ServerAliveInterval 30 in the OpenSSH config, matching PuTTY's keepalive.
StrictHostKeyChecking left on — silencing it defeats the purpose.
A per-environment key rather than one key for everything.

Questions people ask about this

Which client does Git for Windows use?

It depends on the GIT_SSH environment variable — either its bundled OpenSSH or Plink from PuTTY. Check that variable when Git authentication behaves differently than your terminal.

Can I use one key file with both clients?

Yes. Keep the OpenSSH format as the original and export a .ppk copy from PuTTYgen.

Independent guide. Software names belong to their respective developers; always download from the vendor's own site. Back to all Windows SSH guides.