ssh-agent is a program to hold private keys used for public key authentication (RSA, DSA). ssh-agent is often started at the beginning of a login session. All other windows or programs are started as clients to the ssh-agent program. Through use of environment variables, the agent can be located and automatically used for authentication when logging in to other machines using ssh(1). (See System Administration Guide: Security Services.)
If a command line is given, this is executed as a subprocess of the agent. When the command dies, so does the agent.
The agent initially does not have any private keys. Keys are added using ssh-add(1), which sends the identity to the agent. Several identities can be stored in the agent; the agent can
automatically use any of these identities. Use the -l option in ssh-add(1) to display the identities currently held by the agent.
The agent is run in the user's local host. Authentication data need not be stored on any other machine, and authentication passphrases never go over the network. However, if the connection to the agent is forwarded over SSH remote logins, the user can use the privileges given
by the identities anywhere in the network in a secure way.
There are two main ways to get an agent setup. Either you let the agent start a new subcommand into which some environment variables are exported, or you let the agent print the needed shell commands (either sh(1) or csh(1) syntax can be generated) which can be evalled in the calling shell. Later, use ssh(1) to look at these variables and use them to establish a connection to the agent.
A unix-domain socket is created (/tmp/ssh-XXXXXXXX/agent.pid) and the name of this socket is stored in the SSH_AUTH_SOCK environment variable. The socket is made accessible only to the current user. This method is easily abused by root
or another instance of the same user.
The SSH_AGENT_PID environment variable holds the agent's PID.
The agent exits automatically when the command given on the command line terminates.