1 .\" $OpenBSD: ssh-agent.1,v 1.8 1999/11/16 23:59:21 deraadt Exp $
7 .\" Author: Tatu Ylonen <ylo@cs.hut.fi>
9 .\" Copyright (c) 1995 Tatu Ylonen <ylo@cs.hut.fi>, Espoo, Finland
10 .\" All rights reserved
12 .\" Created: Sat Apr 23 20:10:43 1995 ylo
14 .Dd September 25, 1999
19 .Nd authentication agent
30 is a program to hold authentication private keys. The
33 is started in the beginning of an X-session or a login session, and
34 all other windows or programs are started as clients to the ssh-agent
35 program. Through use of environment variables the agent can be located
36 and automatically used for RSA authentication when logging in to other
40 The options are as follows:
43 Generate C-shell commands on
45 This is the default if
47 looks like it's a csh style of shell.
49 Generate Bourne shell commands on
51 This is the default if
53 does not look like it's a csh style of shell.
55 Kill the current agent (given by the
57 environment variable).
60 If a commandline is given, this is executed as a subprocess of the agent.
61 When the command dies, so does the agent.
63 The agent initially does not have any private keys. Keys are added
66 When executed without arguments,
69 .Pa $HOME/.ssh/identity
70 file. If the identity has a passphrase,
72 asks for the passphrase (using a small X11 application if running
73 under X11, or from the terminal if running without X). It then sends
74 the identity to the agent. Several identities can be stored in the
75 agent; the agent can automatically use any of these identities.
77 displays the identities currently held by the agent.
79 The idea is that the agent is run in the user's local PC, laptop, or
80 terminal. Authentication data need not be stored on any other
81 machine, and authentication passphrases never go over the network.
82 However, the connection to the agent is forwarded over SSH
83 remote logins, and the user can thus use the privileges given by the
84 identities anywhere in the network in a secure way.
86 There are two main ways to get an agent setup: Either you let the agent
87 start a new subcommand into which some environment variables are exported, or
88 you let the agent print the needed shell commands (either
92 syntax can be generated) which can be evalled in the calling shell.
95 look at these variables and use them to establish a connection to the agent.
97 A unix-domain socket is created
98 .Pq Pa /tmp/ssh-XXXXXXXX/agent.<pid> ,
99 and the name of this socket is stored in the
102 variable. The socket is made accessible only to the current user.
103 This method is easily abused by root or another instance of the same
108 environment variable holds the agent's PID.
110 The agent exits automatically when the command given on the command
114 .It Pa $HOME/.ssh/identity
115 Contains the RSA authentication identity of the user. This file
116 should not be readable by anyone but the user. It is possible to
117 specify a passphrase when generating the key; that passphrase will be
118 used to encrypt the private part of this file. This file
121 but is normally added to the agent using
124 .It Pa /tmp/ssh-XXXX/agent.<pid> ,
125 Unix-domain sockets used to contain the connection to the
126 authentication agent. These sockets should only be readable by the
127 owner. The sockets should get automatically removed when the agent
130 Tatu Ylonen <ylo@cs.hut.fi>
133 is a derivative of the original (free) ssh 1.2.12 release, but with bugs
134 removed and newer features re-added. Rapidly after the 1.2.12 release,
135 newer versions bore successively more restrictive licenses. This version
139 has all components of a restrictive nature (ie. patents, see
141 directly removed from the source code; any licensed or patented components
145 has been updated to support ssh protocol 1.5.
147 contains added support for
149 authentication and ticket passing.
151 supports one-time password authentication with
155 The libraries described in
157 are required for proper operation.