Consult provides search and navigation commands based on the Emacs completion function completing-read. Completion allows you to quickly select an item from a list of candidates. Consult offers asynchronous and interactive consult-grep and consult-ripgrep commands, and the line-based search command consult-line. Furthermore Consult provides an advanced buffer switching command consult-buffer to switch between buffers, recently opened files, bookmarks and buffer-like candidates from other sources. Some of the Consult commands are enhanced versions of built-in Emacs commands. For example the command consult-imenu presents a flat list of the Imenu with live preview, grouping and narrowing. Please take a look at the full list of commands.

Consult is fully compatible with completion systems centered around the standard Emacs completing-read API, notably the default completion system, Vertico, Mct, and Icomplete.

This package keeps the completion system specifics to a minimum. The ability of the Consult commands to work well with arbitrary completion systems is one of the main advantages of the package. Consult fits well into existing setups and it helps you to create a full completion environment out of small and independent components.

You can combine the complementary packages Marginalia, Embark and Orderless with Consult. Marginalia enriches the completion display with annotations, e.g., documentation strings or file information. The versatile Embark package provides local actions, comparable to a context menu. These actions operate on the selected candidate in the minibuffer or at point in normal buffers. For example, when selecting from a list of files, Embark offers an action to delete the file. Additionally Embark offers a facility to collect completion candidates in a collect buffer. The section Embark integration documents in detail how Consult and Embark work together.

Fig. 1: Command consult-git-grep

Fig. 2: Command consult-imenu

Fig. 3: Command consult-line

Most Consult commands follow the meaningful naming scheme consult-<thing>. Many commands implement a little known but convenient Emacs feature called “future history”, which guesses what input the user wants. At a command prompt type M-n and typically Consult will insert the symbol or thing at point into the input.

TIP: If you have Marginalia annotators activated, type M-x ^consult to see all Consult commands with their abbreviated description. Alternatively, type C-h a ^consult to get an overview of all Consult variables and functions with their descriptions.

• consult-buffer: Enhanced version of switch-to-buffer with support for virtual buffers. Supports live preview of buffers and narrowing to the virtual buffer types. You can type f SPC in order to narrow to recent files. Press SPC to show ephemeral buffers. Supported narrowing keys:
  • b Buffers
  • SPC Hidden buffers
  • * Modified buffers
  • f Files (Requires recentf-mode)
  • r File registers
  • m Bookmarks
  • p Project
  • Custom other sources configured in consult-buffer-sources.
• consult-buffer-other-window, consult-buffer-other-frame, consult-buffer-other-tab: Variants of consult-buffer.
• consult-project-buffer: Variant of consult-buffer restricted to buffers and recent files of the current project. You can add custom sources to consult-project-buffer-sources. The command may prompt you for a project if you invoke it from outside a project.
• consult-bookmark: Select or create bookmark. To select bookmarks you might use the consult-buffer as an alternative, which can include a bookmark virtual buffer source. Note that consult-bookmark supports preview of bookmarks and narrowing.
• consult-recent-file: Select from recent files with preview. You might prefer the powerful consult-buffer instead, which can include recent files as a virtual buffer source. The recentf-mode enables tracking of recent files.

• consult-yank-from-kill-ring: Enhanced version of yank to select an item from the kill-ring. The selected text previewed as overlay in the buffer.
• consult-yank-pop: Enhanced version of yank-pop with DWIM-behavior, which either replaces the last yank by cycling through the kill-ring, or if there has not been a last yank consults the kill-ring. The selected text previewed as overlay in the buffer.
• consult-yank-replace: Like consult-yank-pop, but always replaces the last yank with an item from the kill-ring.
• consult-kmacro: Select macro from the macro ring and execute it.

• consult-register: Select from list of registers. The command supports narrowing to register types and preview of marker positions. This command is useful to search the register contents. For quick access use the commands consult-register-load, consult-register-store or the built-in Emacs register commands.
• consult-register-format: Set register-preview-function to this function for an enhanced register formatting. See the example configuration.
• consult-register-window: Replace register-preview with this function for a better register window. See the example configuration.
• consult-register-load: Utility command to quickly load a register. The command either jumps to the register value or inserts it.
• consult-register-store: Improved UI to store registers depending on the current context with an action menu. With an active region, store/append/prepend the contents, optionally deleting the region when a prefix argument is given. With a numeric prefix argument, store/add the number. Otherwise store point, frameset, window or kmacro. Usage examples:
  • M-' x: If no region is active, store point in register x. If a region is active, store the region in register x.
  • M-' M-w x: Store window configuration in register x.
  • C-u 100 M-' x: Store number in register x.

• consult-goto-line: Jump to line number enhanced with live preview. This is a drop-in replacement for goto-line. Enter a line number to jump to the first column of the given line. Alternatively enter line:column in order to jump to a specific column.
• consult-mark: Jump to a marker in the mark-ring. Supports live preview and recursive editing.
• consult-global-mark: Jump to a marker in the global-mark-ring. Supports live preview and recursive editing.
• consult-outline: Jump to a heading of the outline. Supports narrowing to a heading level, live preview and recursive editing.
• consult-imenu: Jump to imenu item in the current buffer. Supports live preview, recursive editing and narrowing.
• consult-imenu-multi: Jump to imenu item in project buffers, with the same major mode as the current buffer. Supports live preview, recursive editing and narrowing. This feature has been inspired by imenu-anywhere.

• consult-line: Enter search string and select from matching lines. Supports live preview and recursive editing. The symbol at point and the recent Isearch string are added to the “future history” and can be accessed by pressing M-n. When consult-line is bound to the isearch-mode-map and is invoked during a running Isearch, it will use the current Isearch string.
• consult-line-multi: Search dynamically across multiple buffers. By default search across project buffers. If invoked with a prefix argument search across all buffers. The candidates are computed on demand based on the input. The command behaves like consult-grep, but operates on buffers instead of files.
• consult-keep-lines: Replacement for keep/flush-lines which uses the current completion style for filtering the buffer. The function updates the buffer while typing. In particular consult-keep-lines can narrow down an exported Embark collect buffer further, relying on the same completion filtering as completing-read. If the input begins with the negation operator, i.e., ! SPC, the filter matches the complement. If a region is active, the region restricts the filtering.
• consult-focus-lines: Temporarily hide lines by filtering them using the current completion style. Call with C-u prefix argument in order to show the hidden lines again. If the input begins with the negation operator, i.e., ! SPC, the filter matches the complement. In contrast to consult-keep-lines this function does not edit the buffer. If a region is active, the region restricts the filtering.

• consult-grep, consult-ripgrep, consult-git-grep: Search for regular expression in files. Consult invokes Grep asynchronously, while you enter the search term. After at least consult-async-min-input characters, the search gets started. Consult splits the input string into two parts, if the first character is a punctuation character, like #. For example #regexps#filter-string, is split at the second #. The string regexps is passed to Grep. Note that Consult transforms Emacs regular expressions to expressions understand by the search program. Always use Emacs regular expressions at the prompt. If you enter multiple regular expressions separated by space only lines matching all regular expressions are shown. In order to match space literally, escape the space with a backslash. The filter-string is passed to the fast Emacs filtering to further narrow down the list of matches. This is particularly useful if you are using an advanced completion style like orderless. consult-grep supports preview. consult-grep searches the current project directory if a project is found. Otherwise the default-directory is searched. If consult-grep is invoked with prefix argument C-u M-s g, you can specify one or more comma-separated files and directories manually. If invoked with two prefix arguments C-u C-u M-s g, you can first select a project if you are not yet inside a project.
• consult-find, consult-fd, consult-locate: Find file by matching the path against a regexp. Like for consult-grep, either the project root or the current directory is the root directory for the search. The input string is treated similarly to consult-grep, where the first part is passed to find, and the second part is used for Emacs filtering. Prefix arguments to consult-find work just like those for the consult grep commands.

• consult-compile-error: Jump to a compilation error. Supports live preview narrowing and recursive editing.
• consult-flymake: Jump to Flymake diagnostic. Supports live preview and recursive editing. The command supports narrowing. Press e SPC, w SPC, n SPC to only show errors, warnings and notes respectively.
• consult-xref: Integration with xref. This function can be set as xref-show-xrefs-function and xref-show-definitions-function.

• consult-complex-command: Select a command from the command-history. This command is a completing-read version of repeat-complex-command and is also a replacement for the command-history command from chistory.el.
• consult-history: Insert a string from the current buffer history, for example the Eshell or Comint history. You can also invoke this command from the minibuffer. In that case consult-history uses the history stored in the minibuffer-history-variable. If you prefer completion-at-point, take a look at cape-history from the Cape package.
• consult-isearch-history: During an Isearch session, this command picks a search string from history and continues the search with the newly selected string. Outside of Isearch, the command allows you to pick a string from the history and starts a new Isearch. consult-isearch-history acts as a drop-in replacement for isearch-edit-string.

• consult-minor-mode-menu: Enable/disable minor mode. Supports narrowing to on/off/local/global modes by pressing i/o/l/g SPC respectively.
• consult-mode-command: Run a command from the currently active minor or major modes. Supports narrowing to local-minor/global-minor/major mode via the keys l/g/m.

• consult-org-heading: Variant of consult-imenu or consult-outline for Org buffers. The headline and its ancestors headlines are separated by slashes. Supports narrowing by heading level, priority and TODO keyword, as well as live preview and recursive editing.
• consult-org-agenda: Jump to an Org agenda heading. Supports narrowing by heading level, priority and TODO keyword, as well as live preview and recursive editing.

• consult-man: Find Unix man page, via Unix apropos or man -k. consult-man opens the selected man page using the Emacs man command.
• consult-info: Full text search through info pages. If the command is invoked from within an *info* buffer, it will search through the current manual. You may want to create your own commands which search through a predefined set of info pages, for example:

(defun consult-info-emacs ()
  "Search through Emacs info pages."
  (interactive)
  (consult-info "emacs" "efaq" "elisp" "cl" "compat"))

(defun consult-info-org ()
  "Search through the Org info page."
  (interactive)
  (consult-info "org"))

(defun consult-info-completion ()
  "Search through completion info pages."
  (interactive)
  (consult-info "vertico" "consult" "marginalia" "orderless" "embark"
                "corfu" "cape" "tempel"))

• consult-theme: Select a theme and disable all currently enabled themes. Supports live preview of the theme while scrolling through the candidates.
• consult-preview-at-point and consult-preview-at-point-mode: Command and minor mode which previews the candidate at point in the *Completions* buffer. This mode is relevant if you use Mct or the default *Completions* UI.
• consult-completion-in-region: In case you don’t use Corfu as your in-buffer completion UI, this function can be set as completion-in-region-function. Then your minibuffer completion UI (e.g., Vertico or Icomplete) will be used for completion-at-point.

  ;; Use `consult-completion-in-region' if Vertico is enabled.
  ;; Otherwise use the default `completion--in-region' function.
  (setq completion-in-region-function
        (lambda (&rest args)
          (apply (if vertico-mode
                     #'consult-completion-in-region
                   #'completion--in-region)
                 args)))
      

  Instead of consult-completion-in-region, you may prefer to see the completions directly in the buffer as a small popup. In that case, I recommend the Corfu package. There is a technical limitation of consult-completion-in-region in combination with the Lsp modes. The Lsp server relies on the input at point, in order to generate refined candidate strings. Since the completion is transferred from the original buffer to the minibuffer, the server does not receive the updated input. In contrast, in-buffer Lsp completion for example via Corfu works properly since the completion takes place directly in the original buffer.

Consult enhances completing-read with live previews of candidates, additional narrowing capabilities to candidate groups and asynchronously generated candidate lists. The internal consult--read function, which is used by most Consult commands, is a thin wrapper around completing-read and provides the special functionality. In order to support multiple candidate sources there exists the high-level function consult--multi. The architecture of Consult allows it to work with different completion systems in the backend, while still offering advanced features.

Some Consult commands support live previews. For example when you scroll through the items of consult-line, the buffer will scroll to the corresponding position. It is possible to jump back and forth between the minibuffer and the buffer to perform recursive editing while the search is ongoing.

Consult enables previews by default. You can disable them by adjusting the consult-preview-key variable. Furthermore it is possible to specify keybindings which trigger the preview manually as shown in the example configuration. The default setting of consult-preview-key is any which means that Consult triggers the preview immediately on any key press when the selected candidate changes. You can configure each command individually with its own :preview-key. The following settings are possible:

• Automatic and immediate 'any
• Automatic and delayed (list :debounce 0.5 'any)
• Manual and immediate "M-."
• Manual and delayed (list :debounce 0.5 "M-.")
• Disabled nil

A safe recommendation is to leave automatic immediate previews enabled in general and disable the automatic preview only for commands where the preview may be expensive due to file loading. Internally, Consult uses the value of this-command to determine the :preview-key customized. This means that if you wrap a consult-* command within your own function or command, you will also need to add the name of your custom command to the consult-customize call in order for it to be considered.

(consult-customize
 consult-ripgrep consult-git-grep consult-grep
 consult-bookmark consult-recent-file consult-xref
 consult--source-bookmark consult--source-file-register
 consult--source-recent-file consult--source-project-recent-file
 ;; my/command-wrapping-consult    ;; disable auto previews inside my command
 :preview-key '(:debounce 0.4 any) ;; Option 1: Delay preview
 ;; :preview-key "M-.")            ;; Option 2: Manual preview

In this case one may wonder what the difference is between using an Embark action on the current candidate in comparison to a manually triggered preview. The main difference is that the files opened by manual preview are closed again after the completion session. During preview some functionality is disabled to improve the performance, see for example the customization variables consult-preview-variables and consult-preview-allowed-hooks. Only hooks listed in consult-preview-allowed-hooks are executed. This variable applies to find-file-hook, change-major-mode-hook and mode hooks, e.g., prog-mode-hook. In order to enable additional font locking during preview, add the corresponding hooks to the allow list. The following code demonstrates this for org-modern and hl-todo.

;; local modes added to prog-mode hooks
(add-to-list 'consult-preview-allowed-hooks 'hl-todo-mode)
(add-to-list 'consult-preview-allowed-hooks 'elide-head-mode)
;; enabled global modes
(add-to-list 'consult-preview-allowed-hooks 'global-org-modern-mode)
(add-to-list 'consult-preview-allowed-hooks 'global-hl-todo-mode)

Files larger than consult-preview-partial-size are previewed partially. Delaying the preview is also useful for consult-theme, since the theme preview is slow. The delay results in a smoother UI experience.

;; Preview on any key press, but delay 0.5s
(consult-customize consult-theme :preview-key '(:debounce 0.5 any))
;; Preview immediately on M-., on up/down after 0.5s, on any other key after 1s
(consult-customize consult-theme
                   :preview-key
                   '("M-."
                     :debounce 0.5 "<up>" "<down>"
                     :debounce 1 any))

Consult has special support for candidate groups. If the completion UI supports the grouping functionality, the UI separates the groups with thin lines and shows group titles. Grouping is useful if the list of candidates consists of candidates of multiple types or candidates from multiple sources, like the consult-buffer command, which shows both buffers and recently opened files. Note that you can disable the group titles by setting the :group property of the corresponding command to nil using the consult-customize macro.

By entering a narrowing prefix or by pressing a narrowing key it is possible to restrict the completion candidates to a certain candidate group. When you use the consult-buffer command, you can enter the prefix b SPC to restrict list of candidates to buffers only. If you press DEL afterwards, the full candidate list will be shown again. Furthermore a narrowing prefix key and a widening key can be configured which can be pressed to achieve the same effect, see the configuration variables consult-narrow-key and consult-widen-key.

After pressing consult-narrow-key, the possible narrowing keys can be shown by pressing C-h. When pressing C-h after some prefix key, the prefix-help-command is invoked, which shows the keybinding help window by default. As a more compact alternative, there is the consult-narrow-help command which can be bound to a key, for example ? or C-h in the consult-narrow-map, as shown in the example configuration. If which-key is installed, the narrowing keys are automatically shown in the which-key window after pressing the consult-narrow-key.

Consult has support for asynchronous generation of candidate lists. This feature is used for search commands like consult-grep, where the list of matches is generated dynamically while the user is typing a regular expression. The grep process is executed in the background. When modifying the regular expression, the background process is terminated and a new process is started with the modified regular expression.

The matches, which have been found, can then be narrowed using the installed Emacs completion-style. This can be powerful if you are using for example the orderless completion style.

This two-level filtering is possible by splitting the input string. Part of the input string is treated as input to grep and part of the input is used for filtering. There are multiple splitting styles available, configured in consult-async-split-styles-alist: nil, comma, semicolon and perl. The default splitting style is configured with the variable consult-async-split-style.

With the comma and semicolon splitting styles, the first word before the comma or semicolon is passed to grep, the remaining string is used for filtering. The nil splitting style does not perform any splitting, the whole input is passed to grep.

The perl splitting style splits the input string at a punctuation character, using a similar syntax as Perl regular expressions.

Examples:

• #defun: Search for “defun” using grep.
• #consult embark: Search for both “consult” and “embark” using grep in any order.
• #first.*second: Search for “first” followed by “second” using grep.
• #\(consult\|embark\): Search for “consult” or “embark” using grep. Note the usage of Emacs-style regular expressions.
• #defun#consult: Search for “defun” using grep, filter with the word “consult”.
• /defun/consult: It is also possible to use other punctuation characters.
• #to#: Force searching for “to” using grep, since the grep pattern must be longer than consult-async-min-input characters by default.
• #defun -- --invert-match#: Pass argument --invert-match to grep.

Asynchronous processes like find and grep create an error log buffer _*consult-async* (note the leading space), which is useful for troubleshooting. The prompt has a small indicator showing the process status:

• : the usual prompt colon, before input is provided.
• * with warning face, the process is running.
• : with success face, success, process exited with an error code of zero.
• ! with error face, failure, process exited with a nonzero error code.
• ; with error face, interrupted, for example if more input is provided.

Multiple synchronous candidate sources can be combined. This feature is used by the consult-buffer command to present buffer-like candidates in a single menu for quick access. By default consult-buffer includes buffers, bookmarks, recent files and project-specific buffers and files. It is possible to configure the list of sources via the consult-buffer-sources variable. Arbitrary custom sources can be defined.

As an example, the bookmark source is defined as follows:

(defvar consult--source-bookmark
  `(:name     "Bookmark"
    :narrow   ?m
    :category bookmark
    :face     consult-bookmark
    :history  bookmark-history
    :items    ,#'bookmark-all-names
    :action   ,#'consult--bookmark-action))

Required source fields:

• :category Completion category.
• :items List of strings to select from or function returning list of strings. A list of cons cells is not supported.

Optional source fields:

• :name Name of the source, used for narrowing, group titles and annotations.
• :narrow Narrowing character or (character . string) pair.
• :preview-key Preview key or keys which trigger preview.
• :enabled Function which must return t if the source is enabled.
• :hidden When t candidates of this source are hidden by default.
• :face Face used for highlighting the candidates.
• :annotate Annotation function called for each candidate, returns string.
• :history Name of history variable to add selected candidate.
• :default Must be t if the first item of the source is the default value.
• :action Function called with the selected candidate.
• :new Function called with new candidate name, only if :require-match is nil.
• :state State constructor for the source, must return the state function.
• Other source fields can be added specifically to the use case.

The :state and :action fields of the sources deserve a longer explanation. The :action function takes a single argument and is only called after selection with the selected candidate, if the selection has not been aborted. This functionality is provided for convenience and easy definition of sources. The :state field is more general. The :state function is a constructor function without arguments, which can perform some setup necessary for the preview. It must return a closure which takes an ACTION and a CANDIDATE argument. See the docstring of consult--with-preview for more details about the ACTION argument.

By default, consult-buffer previews buffers, bookmarks and files. Loading recent files or bookmarks can result in expensive operations. However it is possible to configure a manual preview as follows.

(consult-customize
 consult--source-bookmark consult--source-file-register
 consult--source-recent-file consult--source-project-recent-file
 :preview-key "M-.")

Sources can be added directly to the consult-buffer-source list for convenience. For example, the following source lists all Org buffers and lets you create new ones.

(defvar org-source
  (list :name     "Org Buffer"
        :category 'buffer
        :narrow   ?o
        :face     'consult-buffer
        :history  'buffer-name-history
        :state    #'consult--buffer-state
        :new
        (lambda (name)
          (with-current-buffer (get-buffer-create name)
            (insert "#+title: " name "\n\n")
            (org-mode)
            (consult--buffer-action (current-buffer))))
        :items
        (lambda ()
          (consult--buffer-query :mode 'org-mode :as #'consult--buffer-pair))))

(add-to-list 'consult-buffer-sources 'org-source 'append)

One can create similar sources for other major modes. See the Consult wiki for many additional source examples. See also the documentation of consult-buffer and of the internal consult--multi API. The function consult--multi can be used to create new multi-source commands.

NOTE: Install the embark-consult package from MELPA, which provides Consult-specific Embark actions and the Occur buffer export.

Embark is a versatile package which offers context dependent actions, comparable to a context menu. See the Embark manual for an extensive description of its capabilities.

Actions are commands which can operate on the currently selected candidate (or target in Embark terminology). When completing files, for example the delete-file command is offered. With Embark you can execute arbitrary commands on the currently selected candidate via M-x.

Furthermore Embark provides the embark-collect command, which collects candidates and presents them in an Embark collect buffer, where further actions can be applied to them. A related feature is the embark-export command, which exports candidate lists to a buffer of a special type. For example in the case of file completion, a Dired buffer is opened.

In the context of Consult, particularly exciting is the possibility to export the matching lines from consult-line, consult-outline, consult-mark and consult-global-mark. The matching lines are exported to an Occur buffer where they can be edited via the occur-edit-mode (press key e). Similarly, Embark supports exporting the matches found by consult-grep, consult-ripgrep and consult-git-grep to a Grep buffer, where the matches across files can be edited, if the wgrep package is installed. These three workflows are symmetric.

• consult-line -> embark-export to occur-mode buffer -> occur-edit-mode for editing of matches in buffer.
• consult-grep -> embark-export to grep-mode buffer -> wgrep for editing of all matches.
• consult-find -> embark-export to dired-mode buffer -> wdired-change-to-wdired-mode for editing.

Consult can be installed from ELPA or MELPA via the Emacs built-in package manager. Alternatively it can be directly installed from the development repository via other non-standard package managers.

There is the Consult wiki, where additional configuration examples can be contributed.

IMPORTANT: It is recommended that you enable lexical binding in your configuration. Many Consult-related code snippets require lexical binding, since they use lambdas and closures.

The Consult package only provides commands and does not add any keybindings or modes. Therefore the package is non-intrusive but requires a little setup effort. In order to use the Consult commands, it is advised to add keybindings for commands which are accessed often. Rarely used commands can be invoked via M-x. Feel free to only bind the commands you consider useful to your workflow. The configuration shown here relies on the use-package macro, which is a convenient tool to manage package configurations.

NOTE: There is the Consult wiki, where you can contribute additional configuration examples.

;; Example configuration for Consult
(use-package consult
  ;; Replace bindings. Lazily loaded by `use-package'.
  :bind (;; C-c bindings in `mode-specific-map'
         ("C-c M-x" . consult-mode-command)
         ("C-c h" . consult-history)
         ("C-c k" . consult-kmacro)
         ("C-c m" . consult-man)
         ("C-c i" . consult-info)
         ([remap Info-search] . consult-info)
         ;; C-x bindings in `ctl-x-map'
         ("C-x M-:" . consult-complex-command)     ;; orig. repeat-complex-command
         ("C-x b" . consult-buffer)                ;; orig. switch-to-buffer
         ("C-x 4 b" . consult-buffer-other-window) ;; orig. switch-to-buffer-other-window
         ("C-x 5 b" . consult-buffer-other-frame)  ;; orig. switch-to-buffer-other-frame
         ("C-x t b" . consult-buffer-other-tab)    ;; orig. switch-to-buffer-other-tab
         ("C-x r b" . consult-bookmark)            ;; orig. bookmark-jump
         ("C-x p b" . consult-project-buffer)      ;; orig. project-switch-to-buffer
         ;; Custom M-# bindings for fast register access
         ("M-#" . consult-register-load)
         ("M-'" . consult-register-store)          ;; orig. abbrev-prefix-mark (unrelated)
         ("C-M-#" . consult-register)
         ;; Other custom bindings
         ("M-y" . consult-yank-pop)                ;; orig. yank-pop
         ;; M-g bindings in `goto-map'
         ("M-g e" . consult-compile-error)
         ("M-g f" . consult-flymake)               ;; Alternative: consult-flycheck
         ("M-g g" . consult-goto-line)             ;; orig. goto-line
         ("M-g M-g" . consult-goto-line)           ;; orig. goto-line
         ("M-g o" . consult-outline)               ;; Alternative: consult-org-heading
         ("M-g m" . consult-mark)
         ("M-g k" . consult-global-mark)
         ("M-g i" . consult-imenu)
         ("M-g I" . consult-imenu-multi)
         ;; M-s bindings in `search-map'
         ("M-s d" . consult-find)                  ;; Alternative: consult-fd
         ("M-s c" . consult-locate)
         ("M-s g" . consult-grep)
         ("M-s G" . consult-git-grep)
         ("M-s r" . consult-ripgrep)
         ("M-s l" . consult-line)
         ("M-s L" . consult-line-multi)
         ("M-s k" . consult-keep-lines)
         ("M-s u" . consult-focus-lines)
         ;; Isearch integration
         ("M-s e" . consult-isearch-history)
         :map isearch-mode-map
         ("M-e" . consult-isearch-history)         ;; orig. isearch-edit-string
         ("M-s e" . consult-isearch-history)       ;; orig. isearch-edit-string
         ("M-s l" . consult-line)                  ;; needed by consult-line to detect isearch
         ("M-s L" . consult-line-multi)            ;; needed by consult-line to detect isearch
         ;; Minibuffer history
         :map minibuffer-local-map
         ("M-s" . consult-history)                 ;; orig. next-matching-history-element
         ("M-r" . consult-history))                ;; orig. previous-matching-history-element

  ;; Enable automatic preview at point in the *Completions* buffer. This is
  ;; relevant when you use the default completion UI.
  :hook (completion-list-mode . consult-preview-at-point-mode)

  ;; The :init configuration is always executed (Not lazy)
  :init

  ;; Optionally configure the register formatting. This improves the register
  ;; preview for `consult-register', `consult-register-load',
  ;; `consult-register-store' and the Emacs built-ins.
  (setq register-preview-delay 0.5
        register-preview-function #'consult-register-format)

  ;; Optionally tweak the register preview window.
  ;; This adds thin lines, sorting and hides the mode line of the window.
  (advice-add #'register-preview :override #'consult-register-window)

  ;; Use Consult to select xref locations with preview
  (setq xref-show-xrefs-function #'consult-xref
        xref-show-definitions-function #'consult-xref)

  ;; Configure other variables and modes in the :config section,
  ;; after lazily loading the package.
  :config

  ;; Optionally configure preview. The default value
  ;; is 'any, such that any key triggers the preview.
  ;; (setq consult-preview-key 'any)
  ;; (setq consult-preview-key "M-.")
  ;; (setq consult-preview-key '("S-<down>" "S-<up>"))
  ;; For some commands and buffer sources it is useful to configure the
  ;; :preview-key on a per-command basis using the `consult-customize' macro.
  (consult-customize
   consult-theme :preview-key '(:debounce 0.2 any)
   consult-ripgrep consult-git-grep consult-grep
   consult-bookmark consult-recent-file consult-xref
   consult--source-bookmark consult--source-file-register
   consult--source-recent-file consult--source-project-recent-file
   ;; :preview-key "M-."
   :preview-key '(:debounce 0.4 any))

  ;; Optionally configure the narrowing key.
  ;; Both < and C-+ work reasonably well.
  (setq consult-narrow-key "<") ;; "C-+"

  ;; Optionally make narrowing help available in the minibuffer.
  ;; You may want to use `embark-prefix-help-command' or which-key instead.
  ;; (keymap-set consult-narrow-map (concat consult-narrow-key " ?") #'consult-narrow-help)
)

TIP: If you have Marginalia installed, type M-x customize-variable RET ^consult to see all Consult-specific customizable variables with their current values and abbreviated description. Alternatively, type C-h a ^consult to get an overview of all Consult variables and functions with their descriptions.

Multiple Consult search commands like consult-grep try to discover the current project and search in the project top level directory by default, if a project is found. Otherwise they fall back to the default-directory. By default, Consult uses the Emacs built-in project discovery support (project-current and project-root). It is possible to configure alternative methods via the customization variable consult-project-function.

;; Optionally configure a different project root function.
;; 1. project.el (the default)
(setq consult-project-function #'consult--default-project--function)
;; 2. vc.el (vc-root-dir)
(setq consult-project-function (lambda (_) (vc-root-dir)))
;; 3. locate-dominating-file
(setq consult-project-function (lambda (_) (locate-dominating-file "." ".git")))
;; 4. projectile.el (projectile-project-root)
(autoload 'projectile-project-root "projectile")
(setq consult-project-function (lambda (_) (projectile-project-root)))
;; 5. Disable project support
(setq consult-project-function nil)

NOTE: Consult supports fine-grained customization of individual commands. This configuration feature exists for experienced users with special requirements. There is the Consult wiki, where we collect further configuration examples.

Commands and buffer sources allow flexible, individual customization by using the consult-customize macro. You can override any option passed to the internal consult--read API. Note that since consult--read is part of the internal API, options could be removed, replaced or renamed in future versions of the package.

Useful options are:

• :prompt set the prompt string
• :preview-key set the preview key, default is consult-preview-key
• :initial set the initial input
• :default set the default value
• :history set the history variable symbol
• :add-history add items to the future history, for example symbol at point
• :sort enable or disable sorting
• :group set to nil to disable candidate grouping and titles.
• :inherit-input-method set to non-nil to inherit the input method.

(consult-customize
 ;; Disable preview for `consult-theme' completely.
 consult-theme :preview-key nil
 ;; Set preview for `consult-buffer' to key `M-.'
 consult-buffer :preview-key "M-."
 ;; For `consult-line' change the prompt and specify multiple preview
 ;; keybindings. Note that you should bind <S-up> and <S-down> in the
 ;; `minibuffer-local-completion-map' or `vertico-map' to the commands which
 ;; select the previous or next candidate.
 consult-line :prompt "Search: "
 :preview-key '("S-<down>" "S-<up>"))

The configuration values are evaluated at runtime, just before the completion session is started. Therefore you can use for example thing-at-point to adjust the initial input or the future history.

(consult-customize
 consult-line
 :add-history (seq-some #'thing-at-point '(region symbol)))

(defalias 'consult-line-thing-at-point 'consult-line)

(consult-customize
 consult-line-thing-at-point
 :initial (thing-at-point 'symbol))

Generally it is possible to modify commands for your individual needs by the following techniques:

1. Use consult-customize in order to change the command or source settings.
2. Create your own wrapper function which passes modified arguments to the Consult functions.
3. Create your own buffer multi sources for consult-buffer.
4. Create advices to modify some internal behavior.
5. Write or propose a patch.

I use and recommend this combination of packages:

• consult: This package
• vertico: Fast and minimal vertical completion system
• marginalia: Annotations for the completion candidates
• embark and embark-consult: Action commands, which can act on the completion candidates
• orderless: Completion style which offers flexible candidate filtering
• wgrep: Editing of grep buffers. Use with consult-grep via embark-export.

There exist multiple fine completion UIs beside Vertico, which are supported by Consult. Give them a try and find out which interaction model fits best for you.

• The builtin completion UI, which pops up the *Completions* buffer.
• The builtin icomplete-vertical-mode.
• mct by Protesilaos Stavrou: Minibuffer and Completions in Tandem, which builds on the default completion UI.

Note that all packages are independent and can be exchanged with alternative components, since there exist no hard dependencies. Furthermore it is possible to get started with only default completion and Consult and add more components later to the mix. For example you can omit Marginalia if you don’t need annotations. I highly recommend the Embark package, but in order to familiarize yourself with the other components, you can first start without it - or you could use with Embark right away and add the other components later on.

We document a list of auxiliary packages in the Consult wiki. These packages integrate Consult with special programs or with other packages in the wider Emacs ecosystem.

If you find a bug or suspect that there is a problem with Consult, please carry out the following steps:

1. Search through the issue tracker if your issue has been reported before (and has been resolved eventually) in the meantime.
2. Remove all packages involved in the suspected bug from your installation.
3. Reinstall the newest version of all relevant packages. Updating alone is not sufficient, since package.el sometimes causes miscompilation. The list of packages includes Consult, Compat, Vertico or other completion UIs, Marginalia, Embark and Orderless.
4. Either use the default completion UI or ensure that exactly one of vertico-mode, mct-mode, or icomplete-mode is enabled. The unsupported modes selectrum-mode, ivy-mode, helm-mode, ido-mode and ido-ubiquitous-mode must be disabled.
5. Ensure that the completion-styles variable is properly configured. Try to set completion-styles to a list including substring or orderless.
6. Try to reproduce the issue with the newest stable Emacs version. Start a bare bone Emacs instance with emacs -Q on the command line. Execute the following minimal code snippets in the scratch buffer. This way we can exclude side effects due to configuration settings. If other packages are relevant to reproduce the issue, include them in the minimal configuration snippet.

Minimal setup with Vertico for emacs -Q:

(package-initialize)
(require 'consult)
(require 'vertico)
(vertico-mode)
(setq completion-styles '(substring basic))

Minimal setup with the default completion system for emacs -Q:

(package-initialize)
(require 'consult)
(setq completion-styles '(substring basic))

Please provide the necessary important information with your bug report:

• The minimal configuration snippet used to reproduce the issue.
• Your completion UI (Default completion, Vertico, Mct or Icomplete).
• A stack trace in case the bug triggers an exception.
• Your Emacs version, since bugs may be fixed or introduced in newer versions.
• Your operating system, since Emacs behavior varies subtly between Linux, Mac and Windows.
• The package manager, e.g., straight.el or package.el, used to install the Emacs packages, in order to exclude update issues. Did you install Consult as part of the Doom Emacs distribution?
• Do you use Evil? Consult does not provide Evil integration out of the box, but there is some support in evil-collection.

When evaluating Consult-related code snippets you should enable lexical binding. Consult often relies on lambdas and lexical closures.

If you have a completion source that’s both dynamic and expensive to generate, completing-read may not be the best choice. Instead, consult--read serves as a thin wrapper around completing-read that provides this functionality. For example, consider the following slow script that splits its input on space:

#!/usr/bin/env bash
# simulate work
sleep .1
# generate completion candidates
printf "%s\n" "$*" | tr " " "\n" | sort

Let’s assume this script is callable as testibus hello world. To have Consult use it for completion, use consult--async-command:

(consult--read
 (consult--async-command (lambda (input) (list "testibus" (string-trim input))))
 :prompt "run testibus: ")

If the completion candidates are generated by Lisp instead, use consult--dynamic-collection:

(consult--read
 (consult--dynamic-collection
  (lambda (input)
    (sleep-for 0.1) ;; Simulate work
    (sort (split-string (string-trim input) nil t) #'string<)))
 :prompt "run testibus: ")

Implementing live preview requires the definition of a state or preview function as defined by consult--with-preview. The preview function receives the candidate and some action to perform (e.g., 'preview). In its simplest form supporting live preview, it looks something like this:

(defun testibus--preview (action cand)
  (pcase action
    ('preview
     (with-current-buffer-window " *testibus*" 'action nil
       (erase-buffer)
       (insert (format "input: %s\n" cand))))))

See the docstring of consult--with-preview for the lifecycle of the action argument. Once defined, we can use this preview function in consult--read:

(consult--read
 (consult--dynamic-collection
  (lambda (input)
    (sleep-for 0.1) ;; Simulate work
    (sort (split-string (string-trim input) nil t) #'string<)))
 :prompt "run testibus: "
 :state #'testibus--preview)

Consult is a community effort, please participate in the discussions. Contributions are welcome, but you may want to discuss potential contributions first. Since this package is part of GNU ELPA contributions require a copyright assignment to the FSF.

If you have a proposal, take a look at the Consult issue tracker and the Consult wishlist. There have been many prior feature discussions. Please search through the issue tracker, maybe your issue or feature request has already been discussed. You can contribute to the Consult wiki, in case you want to share small configuration or command snippets.

This package took inspiration from Counsel by Oleh Krehel. Some of the Consult commands originated in the Counsel package or the wiki of the Selectrum package. This package exists only thanks to the help of these great contributors and thanks to the feedback of many users. Thank you!

Code contributions: Aymeric Agon-Rambosson, Amos Bird, Ashton Wiersdorf, Adam Spiers, Augusto Stoffel, Clemens Radermacher, Zhengyi, Geoffrey Lessel, Illia Ostapyshyn, jakanakaevangeli, JD Smith, Jean-Philippe Bernardy, mattiasdrp, Mohamed Abdelnour, Mohsin Kaleem, Fox Kiester, Omar Antolín Camarena, Earl Hyatt, Omar Polo, Piotr Kwiecinski, Robert Weiner, Sergey Kostyaev, Alexandru Scvorțov, Tecosaur, Sylvain Rousseau, Tom Fitzhenry, Iñigo Serna and Alex Kreisher.

Advice and useful discussions: Enrique Kessler Martínez, Adam Porter, Bruce d’Arcus, Clemens Radermacher, Dmitry Gutov, Howard Melman, Itai Y. Efrat, JD Smith, Manuel Uberti, Stefan Monnier, Omar Antolín Camarena, Steve Purcell, Radon Rosborough, Tom Fitzhenry and Protesilaos Stavrou.