The options are as follows:
The allow-remote-pkcs11 option allows clients of a forwarded ssh-agent to load PKCS#11 or FIDO provider libraries. By default only local clients may perform this operation. Note that signalling that an ssh-agent client is remote is performed by ssh(1), and use of other tools to forward access to the agent socket may circumvent this restriction.
The no-restrict-websafe option instructs ssh-agent to permit signatures using FIDO keys that might be web authentication requests. By default, ssh-agent refuses signature requests for FIDO keys where the key application string does not start with ``ssh:'' and when the data to be signed does not appear to be a ssh(1) user authentication request or a ssh-keygen1 signature. The default behaviour prevents forwarded access to a FIDO key from also implicitly forwarding the ability to authenticate to websites.
Alternately the websafe-allow option allows specifying a pattern-list of key application strings to replace the default application allow-list, for example: ``websafe-allow=ssh:*,example.org,*.example.com''
See PATTERNS in ssh_config5 for a description of pattern-list syntax.
See PATTERNS in ssh_config5 for a description of pattern-list syntax.
There are three main ways to get an agent set up. The first is at the start of an X session, where all other windows or programs are started as children of the ssh-agent program. The agent starts a command under which its environment variables are exported, for example ssh-agent xterm & When the command terminates, so does the agent.
The second method is used for a login session. When ssh-agent is started, it prints the shell commands required to set its environment variables, which in turn can be evaluated in the calling shell, for example eval `ssh-agent -s`
In both of these cases, ssh(1) looks at these environment variables and uses them to establish a connection to the agent.
The third way to run ssh-agent is via socket activation from a supervising process, such as systemd. In this mode, the supervising process creates the listening socket and is responsible for starting ssh-agent as needed, and also for communicating the location of the socket listener to other programs in the user's session. Socket activation is used when ssh-agent is started with either of the -d or -D flags, no socket listening address specified by the -a flag, and both the LISTEN_FDS and LISTEN_PID environment variables correctly supplied by the supervising process.
The agent initially does not have any private keys. Keys are added using ssh-add1 or by ssh(1) when AddKeysToAgent is set in ssh_config5. Multiple identities may be stored in ssh-agent concurrently and ssh(1) will automatically use them if present. ssh-add1 is also used to remove keys from ssh-agent and to query the keys that are held in one.
Connections to ssh-agent may be forwarded from further remote hosts using the -A option to ssh(1) (but see the caveats documented therein), avoiding the need for authentication data to be stored on other machines. Authentication passphrases and private keys never go over the network: the connection to the agent is forwarded over SSH remote connections and the result is returned to the requester, allowing the user access to their identities anywhere in the network in a secure fashion.
ssh-agent will delete all keys it has loaded upon receiving SIGUSR1
In Debian, ssh-agent is installed with the set-group-id bit set, to prevent ptrace(2) attacks retrieving private key material. This has the side-effect of causing the run-time linker to remove certain environment variables which might have security implications for set-id programs, including LD_PRELOAD LD_LIBRARY_PATH and TMPDIR If you need to set any of these environment variables, you will need to do so in the program executed by ssh-agent.