Chapter 3. Configuration

While the default configuration (or “preferences”) make Mutt usable right out of the box, it is often desirable to tailor Mutt to suit your own tastes. When Mutt is first invoked, it will attempt to read the “system” configuration file (defaults set by your local system administrator), unless the “-n” command line option is specified. This file is typically /usr/local/share/mutt/Muttrc or /etc/Muttrc. Mutt will next look for a file named .muttrc in your home directory. If this file does not exist and your home directory has a subdirectory named .mutt, Mutt tries to load a file named .mutt/muttrc.

.muttrc is the file where you will usually place your commands to configure Mutt.

In addition, Mutt supports version specific configuration files that are parsed instead of the default files as explained above. For instance, if your system has a Muttrc-0.88 file in the system configuration directory, and you are running version 0.88 of Mutt, this file will be sourced instead of the Muttrc file. The same is true of the user configuration file, if you have a file .muttrc-0.88.6 in your home directory, when you run Mutt version 0.88.6, it will source this file instead of the default .muttrc file. The version number is the same which is visible using the “-v” command line switch or using the show-version key (default: V) from the index menu.

An initialization file consists of a series of commands. Each line of the file may contain one or more commands. When multiple commands are used, they must be separated by a semicolon (“;”).

set realname='Mutt user' ; ignore x-

The hash mark, or pound sign (“#”), is used as a “comment” character. You can use it to annotate your initialization file. All text after the comment character to the end of the line is ignored.

my_hdr X-Disclaimer: Why are you listening to me? # This is a comment

Single quotes (“'”) and double quotes (“"”) can be used to quote strings which contain spaces or other special characters. The difference between the two types of quotes is similar to that of many popular shell programs, namely that a single quote is used to specify a literal string (one that is not interpreted for shell variables or quoting with a backslash [see next paragraph]), while double quotes indicate a string for which should be evaluated. For example, backticks are evaluated inside of double quotes, but not for single quotes.

“\” quotes the next character, just as in shells such as bash and zsh. For example, if want to put quotes “"” inside of a string, you can use “\” to force the next character to be a literal instead of interpreted character.

set realname="Michael \"MuttDude\" Elkins"

“\\” means to insert a literal “\” into the line. “\n” and “\r” have their usual C meanings of linefeed and carriage-return, respectively.

A “\” at the end of a line can be used to split commands over multiple lines as it “escapes” the line end, provided that the split points don't appear in the middle of command names. Lines are first concatenated before interpretation so that a multi-line can be commented by commenting out the first line only.

set status_format="some very \
long value split \
over several lines"

It is also possible to substitute the output of a Unix command in an initialization file. This is accomplished by enclosing the command in backticks (``). In Example 3.5, “Using external command's output in configuration files”, the output of the Unix command “uname -a” will be substituted before the line is parsed. Since initialization files are line oriented, only the first line of output from the Unix command will be substituted.

my_hdr X-Operating-System: `uname -a`

Both environment variables and Mutt variables can be accessed by prepending “$” to the name of the variable. For example,

set record=+sent_on_$HOSTNAME

will cause Mutt to save outgoing messages to a folder named “sent_on_kremvax” if the environment variable $HOSTNAME is set to “kremvax.” (See $record for details.)

Mutt expands the variable when it is assigned, not when it is used. If the value of a variable on the right-hand side of an assignment changes after the assignment, the variable on the left-hand side will not be affected.

The commands understood by Mutt are explained in the next paragraphs. For a complete list, see the command reference.

All configuration files are expected to be in the current locale as specified by the $charset variable which doesn't have a default value since it's determined by Mutt at startup. If a configuration file is not encoded in the same character set the $config_charset variable should be used: all lines starting with the next are recoded from $config_charset to $charset.

This mechanism should be avoided if possible as it has the following implications:

Usage:

Mutt supports grouping addresses logically into named groups. An address or address pattern can appear in several groups at the same time. These groups can be used in patterns (for searching, limiting and tagging) and in hooks by using group patterns. This can be useful to classify mail and take certain actions depending on in what groups the message is. For example, the mutt user's mailing list would fit into the categories “mailing list” and “mutt-related”. Using send-hook, the sender can be set to a dedicated one for writing mailing list messages, and the signature could be set to a mutt-related one for writing to a mutt list — for other lists, the list sender setting still applies but a different signature can be selected. Or, given a group only containing recipients known to accept encrypted mail, “auto-encryption” can be achieved easily.

The group command is used to directly add either addresses or regular expressions to the specified group or groups. The different categories of arguments to the group command can be in any order. The flags -rx and -addr specify what the following strings (that cannot begin with a hyphen) should be interpreted as: either a regular expression or an email address, respectively.

These address groups can also be created implicitly by the alias, lists, subscribe and alternates commands by specifying the optional -group option. For example,

alternates -group me address1 address2
alternates -group me -group work address3

would create a group named “me” which contains all your addresses and a group named “work” which contains only your work address address3. Besides many other possibilities, this could be used to automatically mark your own messages in a mailing list folder as read or use a special signature for work-related messages.

The ungroup command is used to remove addresses or regular expressions from the specified group or groups. The syntax is similar to the group command, however the special character * can be used to empty a group of all of its contents. As soon as a group gets empty because all addresses and regular expressions have been removed, it'll internally be removed, too (i.e. there cannot be an empty group). When removing regular expressions from a group, the pattern must be specified exactly as given to the group command or -group argument.

Usage:

It's usually very cumbersome to remember or type out the address of someone you are communicating with. Mutt allows you to create “aliases” which map a short string to a full address.

The optional -group argument to alias causes the aliased address(es) to be added to the named group.

To remove an alias or aliases (“*” means all aliases):

alias muttdude me@cs.hmc.edu (Michael Elkins)
alias theguys manny, moe, jack

Unlike other mailers, Mutt doesn't require aliases to be defined in a special file. The alias command can appear anywhere in a configuration file, as long as this file is sourced. Consequently, you can have multiple alias files, or you can have all aliases defined in your .muttrc.

On the other hand, the <create-alias> function can use only one file, the one pointed to by the $alias_file variable (which is ~/.muttrc by default). This file is not special either, in the sense that Mutt will happily append aliases to any file, but in order for the new aliases to take effect you need to explicitly source this file too.

source /usr/local/share/Mutt.aliases
source ~/.mail_aliases
set alias_file=~/.mail_aliases

To use aliases, you merely use the alias at any place in Mutt where Mutt prompts for addresses, such as the To: or Cc: prompt. You can also enter aliases in your editor at the appropriate headers if you have the $edit_headers variable set.

In addition, at the various address prompts, you can use the tab character to expand a partial alias to the full alias. If there are multiple matches, Mutt will bring up a menu with the matching aliases. In order to be presented with the full list of aliases, you must hit tab without a partial alias, such as at the beginning of the prompt or after a comma denoting multiple addresses.

In the alias menu, you can select as many aliases as you want with the select-entry key (default: <Return>), and use the exit key (default: q) to return to the address prompt.

Usage:

This command allows you to change the default key bindings (operation invoked when pressing a key).

map specifies in which menu the binding belongs. Multiple maps may be specified by separating them with commas (no additional whitespace is allowed). The currently defined maps are:

key is the key (or key sequence) you wish to bind. To specify a control character, use the sequence \Cx, where x is the letter of the control character (for example, to specify control-A use “\Ca”). Note that the case of x as well as \C is ignored, so that \CA, \Ca, \cA and \ca are all equivalent. An alternative form is to specify the key as a three digit octal number prefixed with a “\” (for example \177 is equivalent to \c?). In addition, key may be a symbolic name as shown in Table 3.1, “Symbolic key names”.

key does not need to be enclosed in quotes unless it contains a space (“ ”) or semi-colon (“;”).

function specifies which action to take when key is pressed. For a complete list of functions, see the reference. Note that the bind expects function to be specified without angle brackets.

The special function <noop> unbinds the specified key sequence.

Usage:

The charset-hook command defines an alias for a character set. This is useful to properly display messages which are tagged with a character set name not known to Mutt.

The iconv-hook command defines a system-specific name for a character set. This is helpful when your systems character conversion library insists on using strange, system-specific names for character sets.

Usage:

It is often desirable to change settings based on which mailbox you are reading. The folder-hook command provides a method by which you can execute any configuration command. regexp is a regular expression specifying in which mailboxes to execute command before loading. If a mailbox matches multiple folder-hooks, they are executed in the order given in the .muttrc.

folder-hook mutt "set sort=threads"

The following example will set the sort variable to date-sent for all folders but to threads for all folders containing “mutt” in their name.

folder-hook . "set sort=date-sent"
folder-hook mutt "set sort=threads"

Usage:

Macros are useful when you would like a single key to perform a series of actions. When you press key in menu menu, Mutt will behave as if you had typed sequence. So if you have a common sequence of commands you type, you can create a macro to execute those commands with a single key or fewer keys.

menu is the map which the macro will be bound in. Multiple maps may be specified by separating multiple menu arguments by commas. Whitespace may not be used in between the menu arguments and the commas separating them.

key and sequence are expanded by the same rules as the key bindings with some additions. The first is that control characters in sequence can also be specified as ^x. In order to get a caret (“^”) you need to use ^^. Secondly, to specify a certain key such as up or to invoke a function directly, you can use the format <key name> and <function name>. For a listing of key names see the section on key bindings. Functions are listed in the reference.

The advantage with using function names directly is that the macros will work regardless of the current key bindings, so they are not dependent on the user having particular key definitions. This makes them more robust and portable, and also facilitates defining of macros in files used by more than one user (e.g., the system Muttrc).

Optionally you can specify a descriptive text after sequence, which is shown in the help screens if they contain a description.

Usage:

If your terminal supports color, you can spice up Mutt by creating your own color scheme. To define the color of an object (type of information), you must specify both a foreground color and a background color (it is not possible to only specify one or the other).

header and body match regexp in the header/body of a message, index matches pattern (see Section 3, “Patterns: Searching, Limiting and Tagging”) in the message index.

object can be one of:

foreground and background can be one of the following:

foreground can optionally be prefixed with the keyword bright to make the foreground color boldfaced (e.g., brightred).

If your terminal supports it, the special keyword default can be used as a transparent color. The value brightdefault is also valid. If Mutt is linked against the S-Lang library, you also need to set the $COLORFGBG environment variable to the default colors of your terminal for this to work; for example (for Bourne-like shells):

set COLORFGBG="green;black"
export COLORFGBG

Mutt also recognizes the keywords color0, color1, ..., colorN-1 (N being the number of colors supported by your terminal). This is useful when you remap the colors for your display (for example by changing the color associated with color2 for your xterm), since color names may then lose their normal meaning.

If your terminal does not support color, it is still possible change the video attributes through the use of the “mono” command. Usage:

For object, see the color command. attribute can be one of the following:

Usage:

With various functions, Mutt will treat messages differently, depending on whether you sent them or whether you received them from someone else. For instance, when replying to a message that you sent to a different party, Mutt will automatically suggest to send the response to the original message's recipients — responding to yourself won't make much sense in many cases. (See $reply_to.)

Many users receive e-mail under a number of different addresses. To fully use Mutt's features here, the program must be able to recognize what e-mail addresses you receive mail under. That's the purpose of the alternates command: It takes a list of regular expressions, each of which can identify an address under which you receive e-mail.

As addresses are matched using regular expressions and not exact strict comparisons, you should make sure you specify your addresses as precise as possible to avoid mismatches. For example, if you specify:

alternates user@example

Mutt will consider “some-user@example” as being your address, too which may not be desired. As a solution, in such cases addresses should be specified as:

alternates '^user@example$'

The -group flag causes all of the subsequent regular expressions to be added to the named group.

The unalternates command can be used to write exceptions to alternates patterns. If an address matches something in an alternates command, but you nonetheless do not think it is from you, you can list a more precise pattern under an unalternates command.

To remove a regular expression from the alternates list, use the unalternates command with exactly the same regexp. Likewise, if the regexp for an alternates command matches an entry on the unalternates list, that unalternates entry will be removed. If the regexp for unalternates is “*”, all entries on alternates will be removed.

Usage:

Mutt has a few nice features for handling mailing lists. In order to take advantage of them, you must specify which addresses belong to mailing lists, and which mailing lists you are subscribed to. Mutt also has limited support for auto-detecting mailing lists: it supports parsing mailto: links in the common List-Post: header which has the same effect as specifying the list address via the lists command (except the group feature). Once you have done this, the <list-reply> function will work for all known lists. Additionally, when you send a message to a subscribed list, Mutt will add a Mail-Followup-To header to tell other users' mail user agents not to send copies of replies to your personal address.

More precisely, Mutt maintains lists of patterns for the addresses of known and subscribed mailing lists. Every subscribed mailing list is known. To mark a mailing list as known, use the list command. To mark it as subscribed, use subscribe.

You can use regular expressions with both commands. To mark all messages sent to a specific bug report's address on Debian's bug tracking system as list mail, for instance, you could say

subscribe [0-9]+.*@bugs.debian.org

as it's often sufficient to just give a portion of the list's e-mail address.

Specify as much of the address as you need to to remove ambiguity. For example, if you've subscribed to the Mutt mailing list, you will receive mail addressed to mutt-users@mutt.org. So, to tell Mutt that this is a mailing list, you could add lists mutt-users@ to your initialization file. To tell Mutt that you are subscribed to it, add subscribe mutt-users to your initialization file instead. If you also happen to get mail from someone whose address is mutt-users@example.com, you could use lists ^mutt-users@mutt\\.org$ or subscribe ^mutt-users@mutt\\.org$ to match only mail from the actual list.

The -group flag adds all of the subsequent regular expressions to the named address group in addition to adding to the specified address list.

The “unlists” command is used to remove a token from the list of known and subscribed mailing-lists. Use “unlists *” to remove all tokens.

To remove a mailing list from the list of subscribed mailing lists, but keep it on the list of known mailing lists, use unsubscribe.

Usage:

This command is used to move read messages from a specified mailbox to a different mailbox automatically when you quit or change folders. pattern is a regular expression specifying the mailbox to treat as a “spool” mailbox and mailbox specifies where mail should be saved when read.

Unlike some of the other hook commands, only the first matching pattern is used (it is not possible to save read mail in more than a single mailbox).

Usage:

This command specifies folders which can receive mail and which will be checked for new messages periodically.

folder can either be a local file or directory (Mbox/Mmdf or Maildir/Mh). If Mutt was built with POP and/or IMAP support, folder can also be a POP/IMAP folder URL. The URL syntax is described in Section 1.2, “URL Syntax”, POP and IMAP are described in Section 3, “POP3 Support” and Section 4, “IMAP Support” respectively.

Mutt provides a number of advanced features for handling (possibly many) folders and new mail within them, please refer to Section 10, “New Mail Detection” for details (including in what situations and how often Mutt checks for new mail).

The “unmailboxes” command is used to remove a token from the list of folders which receive mail. Use “unmailboxes *” to remove all tokens.

Usage:

The my_hdr command allows you to create your own header fields which will be added to every message you send and appear in the editor if $edit_headers is set.

For example, if you would like to add an “Organization:” header field to all of your outgoing messages, you can put the command something like shown in Example 3.11, “Defining custom headers” in your .muttrc.

my_hdr Organization: A Really Big Company, Anytown, USA

If you would like to add a header field to a single message, you should either set the $edit_headers variable, or use the <edit-headers> function (default: “E”) in the compose menu so that you can edit the header of your message along with the body.

To remove user defined header fields, use the unmy_hdr command. You may specify an asterisk (“*”) to remove all header fields, or the fields to remove. For example, to remove all “To” and “Cc” header fields, you could use:

unmy_hdr to cc

Usage:

This command is used to override the default mailbox used when saving messages. mailbox will be used as the default if the message matches pattern, see Message Matching in Hooks for information on the exact format.

To provide more flexibility and good defaults, Mutt applies the expandos of $index_format to mailbox after it was expanded.

# default: save all to ~/Mail/<author name>
save-hook . ~/Mail/%F

# save from me@turing.cs.hmc.edu and me@cs.hmc.edu to $folder/elkins
save-hook me@(turing\\.)?cs\\.hmc\\.edu$ +elkins

# save from aol.com to $folder/spam
save-hook aol\\.com$ +spam

Also see the fcc-save-hook command.

Usage:

This command is used to save outgoing mail in a mailbox other than $record. Mutt searches the initial list of message recipients for the first matching regexp and uses mailbox as the default Fcc: mailbox. If no match is found the message will be saved to $record mailbox.

To provide more flexibility and good defaults, Mutt applies the expandos of $index_format to mailbox after it was expanded.

See Message Matching in Hooks for information on the exact format of pattern.

fcc-hook [@.]aol\\.com$ +spammers

...will save a copy of all messages going to the aol.com domain to the `+spammers' mailbox by default. Also see the fcc-save-hook command.

Usage:

This command is a shortcut, equivalent to doing both a fcc-hook and a save-hook with its arguments, including %-expansion on mailbox according to $index_format.

Usage:

These commands can be used to execute arbitrary configuration commands based upon recipients of the message. pattern is used to match the message, see Message Matching in Hooks for details. command is executed when pattern matches.

reply-hook is matched against the message you are replying to, instead of the message you are sending. send-hook is matched against all messages, both new and replies.

send2-hook is matched every time a message is changed, either by editing it, or by using the compose menu to change its recipients or subject. send2-hook is executed after send-hook, and can, e.g., be used to set parameters such as the $sendmail variable depending on the message's sender address.

For each type of send-hook or reply-hook, when multiple matches occur, commands are executed in the order they are specified in the .muttrc (for that type of hook).

Example: send-hook mutt "set mime_forward signature=''"

Another typical use for this command is to change the values of the $attribution, $signature and $locale variables in order to change the language of the attributions and signatures based upon the recipients.

Usage:

This command can be used to execute arbitrary configuration commands before viewing or formatting a message based upon information about the message. command is executed if the pattern matches the message to be displayed. When multiple matches occur, commands are executed in the order they are specified in the .muttrc.

See Message Matching in Hooks for information on the exact format of pattern.

Example:

message-hook ~A 'set pager=builtin'
message-hook '~f freshmeat-news' 'set pager="less \"+/^  subject: .*\""'

Usage:

When encrypting messages with PGP/GnuPG or OpenSSL, you may want to associate a certain key with a given e-mail address automatically, either because the recipient's public key can't be deduced from the destination address, or because, for some reasons, you need to override the key Mutt would normally use. The crypt-hook command provides a method by which you can specify the ID of the public key to be used when encrypting messages to a certain recipient.

The meaning of keyid is to be taken broadly in this context: You can either put a numerical key ID here, an e-mail address, or even just a real name.

Usage:

This command adds the named string to the keyboard buffer. The string may contain control characters, key names and function names like the sequence string in the macro command. You may use it to automatically run a sequence of commands at startup, or when entering certain folders. For example, Example 3.13, “Embedding push in folder-hook” shows how to automatically collapse all threads when entering a folder.

folder-hook . 'push <collapse-all>'

For using functions like shown in the example, it's important to use angle brackets (“<” and “>”) to make Mutt recognize the input as a function name. Otherwise it will simulate individual just keystrokes, i.e. “push collapse-all” would be interpreted as if you had typed “c”, followed by “o”, followed by “l”, ..., which is not desired and may lead to very unexpected behavior.

Keystrokes can be used, too, but are less portable because of potentially changed key bindings. With default bindings, this is equivalent to the above example:

folder-hook . 'push \eV'

because it simulates that Esc+V was pressed (which is the default binding of <collapse-all>).

Usage:

This command can be used to execute any function. Functions are listed in the function reference. “exec function” is equivalent to “push <function>”.

Usage:

The score commands adds value to a message's score if pattern matches it. pattern is a string in the format described in the patterns section (note: For efficiency reasons, patterns which scan information not available in the index, such as ~b, ~B or ~h, may not be used). value is a positive or negative integer. A message's final score is the sum total of all matching score entries. However, you may optionally prefix value with an equal sign (“=”) to cause evaluation to stop at a particular entry if there is a match. Negative final scores are rounded up to 0.

The unscore command removes score entries from the list. You must specify the same pattern specified in the score command for it to be removed. The pattern “*” is a special token which means to clear the list of all score entries.

Usage:

Mutt has generalized support for external spam-scoring filters. By defining your spam patterns with the spam and nospam commands, you can limit, search, and sort your mail based on its spam attributes, as determined by the external filter. You also can display the spam attributes in your index display using the %H selector in the $index_format variable. (Tip: try %?H?[%H] ? to display spam tags only when they are defined for a given message.)

Your first step is to define your external filter's spam patterns using the spam command. pattern should be a regular expression that matches a header in a mail message. If any message in the mailbox matches this regular expression, it will receive a “spam tag” or “spam attribute” (unless it also matches a nospam pattern — see below.) The appearance of this attribute is entirely up to you, and is governed by the format parameter. format can be any static text, but it also can include back-references from the pattern expression. (A regular expression “back-reference” refers to a sub-expression contained within parentheses.) %1 is replaced with the first back-reference in the regex, %2 with the second, etc.

To match spam tags, mutt needs the corresponding header information which is always the case for local and POP folders but not for IMAP in the default configuration. Depending on the spam header to be analyzed, $imap_headers may need to be adjusted.

If you're using multiple spam filters, a message can have more than one spam-related header. You can define spam patterns for each filter you use. If a message matches two or more of these patterns, and the $spam_separator variable is set to a string, then the message's spam tag will consist of all the format strings joined together, with the value of $spam_separator separating them.

For example, suppose one uses DCC, SpamAssassin, and PureMessage, then the configuration might look like in Example 3.14, “Configuring spam detection”.

spam "X-DCC-.*-Metrics:.*(....)=many"         "90+/DCC-%1"
spam "X-Spam-Status: Yes"                     "90+/SA"
spam "X-PerlMX-Spam: .*Probability=([0-9]+)%" "%1/PM"
set spam_separator=", "

If then a message is received that DCC registered with “many” hits under the “Fuz2” checksum, and that PureMessage registered with a 97% probability of being spam, that message's spam tag would read 90+/DCC-Fuz2, 97/PM. (The four characters before “=many” in a DCC report indicate the checksum used — in this case, “Fuz2”.)

If the $spam_separator variable is unset, then each spam pattern match supersedes the previous one. Instead of getting joined format strings, you'll get only the last one to match.

The spam tag is what will be displayed in the index when you use %H in the $index_format variable. It's also the string that the ~H pattern-matching expression matches against for <search> and <limit> functions. And it's what sorting by spam attribute will use as a sort key.

That's a pretty complicated example, and most people's actual environments will have only one spam filter. The simpler your configuration, the more effective Mutt can be, especially when it comes to sorting.

Generally, when you sort by spam tag, Mutt will sort lexically — that is, by ordering strings alphanumerically. However, if a spam tag begins with a number, Mutt will sort numerically first, and lexically only when two numbers are equal in value. (This is like UNIX's sort -n.) A message with no spam attributes at all — that is, one that didn't match any of your spam patterns — is sorted at lowest priority. Numbers are sorted next, beginning with 0 and ranging upward. Finally, non-numeric strings are sorted, with “a” taking lower priority than “z”. Clearly, in general, sorting by spam tags is most effective when you can coerce your filter to give you a raw number. But in case you can't, Mutt can still do something useful.

The nospam command can be used to write exceptions to spam patterns. If a header pattern matches something in a spam command, but you nonetheless do not want it to receive a spam tag, you can list a more precise pattern under a nospam command.

If the pattern given to nospam is exactly the same as the pattern on an existing spam list entry, the effect will be to remove the entry from the spam list, instead of adding an exception. Likewise, if the pattern for a spam command matches an entry on the nospam list, that nospam entry will be removed. If the pattern for nospam is “*”, all entries on both lists will be removed. This might be the default action if you use spam and nospam in conjunction with a folder-hook.

You can have as many spam or nospam commands as you like. You can even do your own primitive spam detection within Mutt — for example, if you consider all mail from MAILER-DAEMON to be spam, you can use a spam command like this:

spam "^From: .*MAILER-DAEMON"       "999"

Usage:

This command allows the inclusion of initialization commands from other files. For example, I place all of my aliases in ~/.mail_aliases so that I can make my ~/.muttrc readable and keep my aliases private.

If the filename begins with a tilde (“~”), it will be expanded to the path of your home directory.

If the filename ends with a vertical bar (“|”), then filename is considered to be an executable program from which to read input (e.g. source ~/bin/myscript|).

Usage:

This command permits you to flush hooks you have previously defined. You can either remove all hooks by giving the “*” character as an argument, or you can remove all hooks of a specific type by saying something like unhook send-hook.