1. Manuals
  2. Tumbleweed
  3. mupdf
  4. mutool(1)

Scroll to navigation

MUTOOL(1) General Commands Manual MUTOOL(1)

NAME

mutool - all purpose tool for dealing with PDF files

SYNOPSIS

mutool sub-command [options]

DESCRIPTION

mutool is a tool based on MuPDF for dealing with document files in various manners. There are several sub commands available, as described below.

VERSION

mutool -v

Shows the MuPDF version used to build mutool.

AUDIT

mutool audit [options] file1.pdf [file2.pdf ...]

Outputs a report of operator and space usage for each given PDF file.

The output filename for the report.

BAKE

mutool bake [options] input.pdf [output.pdf]

Normally annotations and form field widgets are separate content streams that get drawn on top of the page contents. This command bakes annotations and/or widgets into the page contents, making them inseparable. After baking in annotations and widgets they are just visual representaitons of what the original annotation/widget looked like, their properties can no longer be edited since they no longer exist as proper annotation/widgets.

Do not bake in annotations.
Do not bake in form field widgets.
A comma-separated list of PDF options. See PDF OUTPUT OPTIONS below.

BARCODE

mutool barcode -d [options] file1.pdf [pages1] [file2.pdf] [pages2]

Decodes a barcode from an image or a document.

Use the specified password if the file is encrypted.
The output filename. If no output is specified, any text from decoded barcodes will go to stdout.
-r angle
Hwo much to rotate the input pages in degrees (0-360), before trying to decode any barcodes.
Comma-separated list of page numbers and ranges (for example: 1,5,10-15,20-N), where the character N denotes the last page. If no pages are specified, then all pages will be processed.
mutool barcode -c [options] text

Creates a new barcode containing text argument and formatted according to options specified.

The output format is inferred from the output filename, either PNG or PDF. If no output is specified, the default is out.png
-F format
The desired output barcode format:
aztec
codabar
code39
code93
code128
databar
databarexpanded
datamatrix
ean8
ean13
itf
maxicode
pdf417
qrcode
upca
upce
microqrcode
rmqrcode
dxfilmedge
databarlimited
-s size
Set size in pixels for the output barcode. If not specified the smallest size that can be decoded is chosen.
Add quiet zones around the barcode. This puts an empty margin around created barcodes.
Add human-readable text, when available. Some barcodes, e.g. EAN-13, may have the barcode contents printed in human-readable text next to the barcode, which is enabled by this flag.
-t level
Set error correction level (0-8). Some barcodes, e.g. QR-codes, support several levels of error correction, which can be customized through this option.
The text to be encoded into a barcode.

CLEAN

mutool clean [options] input.pdf [output.pdf] [pages ...]

The clean command pretty prints and rewrites the syntax of a PDF file. It can be used to repair broken files, expand compressed streams, filter out a range of pages, etc.

If no output file is specified, it will write the cleaned PDF to "out.pdf" in the current directory.

Use the specified password if the file is encrypted.
Garbage collect objects that have no references from other objects.
In addition to -g, also renumber all objects and compact the cross reference table.
In addition to -gg, also merge and reuse duplicate objects.
In addition to -ggg, also merge and reuse duplicate streams.
No longer supported! Linearize output. Used to create a "Web Optimized" output file.
Save file without encryption.
Save file with desired encryption method, one of: rc4-40,rc40-128,aes-128, or aes-256.
Owner password used when encrypting output file. This is required to edit the output file.
User password used when encrypting output file. This is required to view the output file.
-P permissions
Permissions used when encrypting output file. These are the permissions that apply when the output file is viewed.
ASCII Hex encode binary streams. Use in conjunction with -d and -i or -f to ensure that although the images and/or fonts are compressed, the resulting file can still be viewed and edited with a text editor.
Decompress streams. This will make the output file larger, but provides easy access for reading and editing the contents with a text editor.
Deflate uncompressed streams.
If combined with -d, any decompressed streams will be recompressed.
If combined with -a, the streams will also be hex encoded after compression.
-e effort
Configure how much effort is spent compressing the output file. 0 is default amount, otherwise a percentage of the maximum effort, 1 for minimum and 100 for maximum effort.
Toggle decompression of font streams. Use in conjunction with -d to leave fonts compressed.
Toggle decompression of image streams. Use in conjunction with -d to leave images compressed.
Pretty-print graphics commands in content streams.
Sanitize graphics commands in content streams, this means, for example, removing invalid or redundant commands.
Use compact PDF object syntax. This saves space.
Use indented PDF object syntax. This is more human-readable.
Print comments containing labels showing how each object can be reached from the Root.
Transform all text into vector graphics commands.
Create appearance streams for annotations that are missing appearance streams.
Recreate appearance streams for all annotations.
Preserve metadata.
Subset fonts if possible. (EXPERIMENTAL!)
Use object streams cross reference streams for extra compressoin.
--(color|gray|bitonal)-(|lossy-|lossless-)image-subsample-method=method
Set the subsampling method (average or bicubic) for the desired image types, for example color-lossy and bitonal-lossless.
--(color|gray|bitonal)-(|lossy-|lossless-)image-subsample-dpi=dpi[,dpi]
Set the resolution at which to subsample.
--(color|gray|bitonal)-(|lossy-|lossless-)image-recompress-method=quality
Set the recompression quality to either of never, same, lossless, jpeg, j2k, fax, or jbig2.
Recompress images when the rewritten images by the subsampling- or recompression methods above are smaller (default), or always recompress them.
Keep or drop the structure tree.
Comma-separated list of page numbers and ranges (for example: 1,5,10-15,20-N), where the character N denotes the last page. If no pages are specified, then all pages will be included.

CONVERT

mutool convert [options] input.pdf [pages ...]

The convert command is used to convert a file from one format to another.

Use the specified password if the file is encrypted.
The output format is inferred from the output filename. Embed %d in the name to indicate the page number (for example: "page%d.png"). Printf modifiers are supported, for example "%03d". If no output is specified, the output will go to stdout.
-F format
Enforce a specific output format. Only necessary when outputting to stdout since normally the output filename is used to infer the output format.
Comma-separated list of format specific output options, see PDF OUTPUT OPTIONS and OUTPUT OPTIONS below.
-b box
Use named page box (MediaBox, CropBox, BleedBox, TrimBox, or ArtBox).
-A bits
Specify how many bits of anti-aliasing to use. The default is 8.
Page width in points for EPUB layout.
Page height in points for EPUB layout.
-S size
Font size in points for EPUB layout.
User CSS stylesheet for EPUB layout. -X Disable document styles for EPUB layout.
Comma-separated list of page numbers and ranges (for example: 1,5,10-15,20-N), where the character N denotes the last page. If no pages are specified, then all pages will be rendered.

CREATE

mutool create [-o output.pdf] [options] page.txt [page2.txt ...]

The create command creates a new PDF file with the contents created from one or more input files containing graphics commands.

If no output file is specified, it will write the created PDF to "out.pdf" in the current directory.
Comma-separated list of format specific output options, see PDF OUTPUT OPTIONS below.
A page is created for each input file, with the contents of the file copied into the content stream. Special comments in the input files are parsed to define the page dimensions and font and image resources:
%%MediaBox LLX LLY URX URY
Set the page size (default A4).
%%Rotate Angle
Set the page rotation (default 90 degrees).
%%Font Name Filename Encoding
Create a simple font resource using the declared Name. Filename can be either a font file or one of the Base-14 font names (Times-Roman, etc). Encoding is one of Latin (for ISO-8859-1), Greek (for ISO-8859-7) or Cyrillic (for KOI-8).
%%Font Name Language WMode Style Language
Create a CID font resource for CJK scripts using the declared Name. Language is one of zh-Hant, zh-Hans, ja, or ko. Wmode is one of H, or V. Style is serif, or sans.
%%Image Im0 path/to/image.png
Create an image resource.
This adds an image, a triangle and some text to a page: 

%%MediaBox 0 0 595 842
%%Font TmRm Times-Roman
%%Font Helv-C Helvetica Cyrillic
%%Font Helv-G Helvetica Greek
%%CJKFont Song zh-Hant
%%CJKFont Mincho ja
%%CJKFont Batang ko
%%Image I0 docs/examples/huntingofthesnark.png

% Draw an image.
q
480 0 0 480 50 250 cm
/I0 Do
Q

% Draw a triangle.
q
1 0 0 rg
50 50 m
100 200 l
200 50 l
f
Q

% Show some text.
q
0 0 1 rg
BT /TmRm 24 Tf 50 760 Td (Hello, world!) Tj ET
BT /Helv-C 24 Tf 50 730 Td <fac4d2c1d7d3d4d7d5cad4c521> Tj ET
BT /Helv-G 24 Tf 50 700 Td <eae1ebe7ecddf1e1> Tj ET
BT /Song 24 Tf 50 670 Td <4F60 597D> Tj ET
BT /Mincho 24 Tf 50 640 Td <3053 3093 306b 3061 306f> Tj ET
BT /Batang 24 Tf 50 610 Td <c548 b155 d558 c138 c694> Tj ET
Q

DRAW

mutool draw [options] file [pages ...]

The draw command will render a document to image files, convert to another vector format, or extract the text content.
The supported input document formats are: pdf, xps, cbz, md, and epub.
The supported output image formats are: pbm, pgm, ppm, pam, png, pwg, pcl, and ps.
The supported output vector formats are: svg,pdf, and debug trace (as xml).
The supported output text formats are: plain text ,html, and structured text (as xml or json).

Use the specified password if the file is encrypted.
The output format is inferred from the output filename. Embed %d in the name to indicate the page number (for example: "page%d.png"). Printf modifiers are supported, for example "%03d". If no output is specified, the output will go to stdout.
-F format
Enforce a specific output format. Only necessary when outputting to stdout since normally the output filename is used to infer the output format.
Be quiet, do not print progress messages.
Rotate clockwise by given number of degrees.
-r resolution
Render the page at the specified resolution. The default resolution is 72 dpi.
Render the page at the specified width (or, if the -r flag is used, render with a maximum width).
Render the page at the specified height (or, if the -r flag is used, render with a maximum height).
Fit exactly; ignore the aspect ratio when matching specified width/heights.
-b box
Use named page box (MediaBox, CropBox, BleedBox, TrimBox, or ArtBox).
Render in banded mode with each band no taller than the given height. This uses less memory during rendering. Only compatible with pam, pgm, ppm, pnm and png output formats. Banded rendering and md5 checksumming may not be used at the same time.
Number of threads to use for rendering (banded mode only).
Page width in points for EPUB layout.
Page height in points for EPUB layout.
-S size
Font size in points for EPUB layout.
User CSS stylesheet for EPUB layout.
Disable document styles for EPUB layout.
Disable usage of accelerator file.
-c colorspace
Render in the specified colorspace. Supported colorspaces are: mono, gray, grayalpha, rgb, rgbalpha, cmyk, cmykalpha. Some abbreviations are allowed: m, g, ga, rgba, cmyka. The default is chosen based on the output format.
-e filename
Proof ICC profile filename to use for rendering.
Apply gamma correction. Some typical values are 0.7 or 1.4 to thin or darken text rendering.
Invert colors.
-s [mft5]
Show various bits of information: m for glyph cache and total memory usage, f for page features such as whether the page is grayscale or color, t for per page rendering times as well statistics, and 5 for md5 checksums of rendered images that can be used to check if rendering has changed.
-A bits
Specify how many bits of anti-aliasing to use. The default is 8.
-A bits/bits
Specify how many bits of anti-aliasing to use. The default is 8. Bits for graphics is before the slash, bits for text after the slash.
-l width
Minimum stroked line width (in pixels).
Do not draw text.
Only draw text.
Disable use of display lists. May cause slowdowns, but should reduce the amount of memory used.
Ignore errors.
-m limit
Limit memory usage in bytes.
Low memory mode (avoid caching objects by clearing cache after each page).
Run interpretation and rendering at the same time.
Disable ICC workflow.
-M overprint
Control spot/overprint rendering: 0 for no spot rendering, 1 for Overprint simulation Overprint simulation (default), or 2 for full spot rendering.
-t language
Specify language/script for OCR (default: eng)
-d ocr-file-path
Specify path for OCR files (default: rely on TESSDATA_PREFIX environment variable.
Set the skew correction, either one of auto, 0 for increase size, 1 for maintain size, or 2 for decrease size.
Set the skew correction as well as the angle.
Print the layer configs to stderr.
Select layer config (by number from -y l ).
Select layer config (by number from -y l ) and toggle the listed items.
Print the individual layers to stderr.
-z layer-number
Hide individual layer.
-Z layer-number
Show individual layer.
Comma-separated list of page numbers and ranges (for example: 1,5,10-15,20-N), where the character N denotes the last page. If no pages are specified, then all pages will be rendered.

EXTRACT

mutool extract [options] file.pdf [object-number ...]

The extract command can be used to extract images and font files from a PDF. If no object numbers are given on the command line, all images and fonts will be extracted.

Use the specified password if the file is encrypted.
Convert images to RGB when extracting them.
Embed SMasks as alpha channel..
Do not use ICC color conversions.

INFO

mutool info [options] file.pdf [pages ...]

The info command lists the resources used on each page in a PDF file. The default is to list all resource types, but if one or more flags are given, only the flagged types will be shown.

Use the specified password if the file is encrypted.
List fonts.
List images.
List page dimensions.
List patterns.
List shadings.
List form and postscript XObjects.
List ZUGFeRD info.
Comma-separated list of page numbers and ranges (for example: 1,5,10-15,20-N), where the character N denotes the last page. If no pages are specified, then all pages will be included.

MERGE

mutool merge [options] file1 [pages1] [file2 [pages2] ...]

The merge command is used to pick out pages from two or more files and merge them in order into a new PDF output file.

The output filename.
Comma-separated list of format specific output options, see PDF OUTPUT OPTIONS below.

PAGES

mutool pages [options] input.pdf [pages ...]

The pages command dumps information about the size and orientation of pages within the document.

Use the specified password if the file is encrypted.
Comma-separated list of page numbers and ranges (for example: 1,5,10-15,20-N), where the character N denotes the last page. If no pages are specified, then all pages will be included.

POSTER

mutool poster [options] input.pdf [output.pdf]

The poster command splits each page into tiles, and puts each tile on a page of its own. It's useful for printing a large page onto smaller pieces of paper that can then be glued together to create a large poster.

Use the specified password if the file is encrypted.
-m margin
Margin (overlap) between pages in points or percent.
Split the page into this many horizontal pieces.
Split the page into this many vertical pieces.
Splits hoirzontally from right to left (default splits from left to right).

The output will have x times y number of pages for each input page.

RECOLOR

mutool recolor [options] input.pdf

Change the color space used in a PDF to a desired one. This command can, for example, change a color PDF to be grayscale.

-c colorspace
Desired output colorspace, choose one of gray (default), rgb, or cmyk.
The output filename.
Remove any output intents.

RUN

mutool run script.js [arguments]

Executes a Javascript program which has access to most of the features of the MuPDF library. The command supports ECMAScript 5 syntax in strict mode. All of the MuPDF constructors and function live in the global object, and the command line arguments are accessible from the global argv object.

If invoke without any arguments, it will drop you into an interactive REPL (read-eval-print-loop). On the interactive prompt, if you prefix a line with an equal character it will automatically print the results of the line.

See the MuPDF documentation at https://mupdf.readthedocs.io for details about the Javascript interfaces.

SHOW

mutool show [options] file.pdf [object-number ...]

The show command will print the specified objects and streams to stdout. Streams are decoded and non-printable characters are represented with a period by default.

Use the specified password if the file is encrypted.
Write output to file instead of stdout.
Print streams in their original encoded (or compressed) form.
Print streams as binary data and omit the object header.
Print each object on a single line, suitable for grep.
Force repair before showing any objects.
Show object labels (how to reach an object from the root).

Specify what to show by using one of the following keywords, or a path to an object:

'trailer'
Print the trailer dictionary.
'xref'
Print the cross reference table.
'pages'
List the object numbers for every page.
'grep'
Print all the objects in the file in a compact one-line format suitable for piping to grep.
'outline'
Print the outline (also known as "table of contents" or "bookmarks").
'js'
Print document level Javascript.
'form'
Print form objects.
<path>
A path starts with either an object number, a property in the trailer dictionary, or the keyword trailerorpages. Separate elements with a period . or slash /. Select a page object by using pages/N where N is the page number. The first page is number 1.
*
You can use * as an element to iterate over all array indices or dictionary properties in an object.

SIGN

mutool sign [options] input.pdf [signature-object-number ...]

Reads a PDF file and prints information about each signature field object. With suitable options all or specified signatures can be verified, cleared or signed.

Use the specified password if the file is encrypted.
Verify all signatures, or only the specified signature.
Clear all signatures, or only the specified signature.
-s certificate-filename
Read a certificate from the given filename and use it to sign all, or just the specified signature.
-P certificate-password
Use the specified password when opening the certificate.
The output filename. signature object numbers A list of PDF object number for signatures to verify/clear/sign. If no such object numbers are given, all signatures will be processed.

TRACE

mutool trace [options] file1 [pages1] [file2 [pages2] ...]

Prints a trace of device calls needed to render the desired pages from the given document.

Use the specified password if the file is encrypted.
-b box
Which box to trim to, either one of: (MediaBox (default), CropBox, BleedBox, TrimBox, or ArtBox).
Page width in points for EPUB layout.
Page height in points for EPUB layout.
-S size
Font size in points for EPUB layout.
User CSS stylesheet for EPUB layout.
Disable document styles for EPUB layout.
Enable use of display lists.
Comma-separated list of page numbers and ranges (for example: 1,5,10-15,20-N), where the character N denotes the last page. If no pages are specified, then all pages will be rendered.

TRIM

mutool trim [options] input.pdf

Makes a modified version of a PDF with content that falls inside (or outside) of the given box removed.

-b box
Which box to trim to, either one of: (MediaBox (default), CropBox, BleedBox, TrimBox, or ArtBox).
Fallback to mediabox if specified box not available.
-m all
Add same margins to all sides of the box, where all is positive for inwards, negative for outwards.
-m vertical,horizontal
Add specific margins for vertical and horizontal sides of the box, where verticalandhorizontal are positive for inwards, negative for outwards.
-m top,right,bottom,left
Add specific margins for top, right, bottom and left sides of the box, where top, right, bottom, and left each are positive for inwards, negative for outwards.
Exclude contents of box rather than include them.
The output filename.

OUTPUT OPTIONS

Configuring output is done by a comma-separated list of output options.

Raster output options

These apply to output formats: cbz, png, pnm, pgm, ppm, pam, pbm, pkm.

Rotate rendered pagess N degrees counterclockwise.
Set both X and Y resolution to N pixels per inch.
Set X resolution of rendred pages in pixels per inch.
Set Y resolution of rendred pages in pixels per inch.
Render pages to fit N pixels wide (ignore resolution option).
Render pages to fit N pixels tall (ignore resolution option).
Render pages using the specified colorspace.
Render pages with alpha channel and transparent background.
Set the rasterizer to use for graphics or text respectively:
Antialias with N bits (0 to 8).
Center of pixel.
Any part of pixel.

PCL options

These apply to print-raster output formats: pcl, pclm, ps, pwg.

Render 1-bit black and white or a full color page.
Render 1-bit black and white or a full color page.
Set vertical spacing to none, PCL 3 (<ESC>*p+<n>Y), PCL 5 (<ESC>*b<n>Y), or PCL 6 (<ESC>*b<n>Y and clear seed row).
Enable mode 2 graphics compression.
Enable mode 3 graphics compression.
End of graphics (<ESC>+rB) resets all parameters.
Duplex supported (<ESC>&l<duplex>S).
Papersize setting supported (<ESC>&l<sizecode>A).
Number of copies supported (<ESC>&l<copies>X).
Disable/enable HP 4PJL model-specific output.
Disable/enable Oce 9050 models-specific output.

PCLm options

These apply only to: pclm.

No compression (default), or enable flate compression.
Strip height (default 16).

PWG options

These apply only to: pwg.

Set the media_class field.
Set the media_color field.
Set the media_type field.
Set the output_type field.
Set the rendering_intent field.
Set the page_size_name field.
Set the advance_distance field.
Set the advance_media field.
Set the collate field.
Set the cut_media field.
Set the duplex field.
Set the insert_sheet field.
Set the jog field.
Set the leading_edge field.
Set the manual_feed field.
Set the media_position field.
Set the media_weight field.
Set the mirror_print field.
Set the negative_print field.
Set the num_copies field.
Set the orientation field.
Set the output_face_up field.
Set the page_size_x field.
Set the page_size_y field.
Set the separations field.
Set the tray_switch field.
Set the tumble field.
Set the media_type_num field.
Set the compression field.
Set the row_count field.
Set the row_feed field.
Set the row_step field.

SVG options

These apply only to: svg.

Emit text as <text> elements (inaccurate fonts).
Emit text as <path> elements (accurate fonts).
Do not reuse image using <symbol> definitions.

Text options

These apply to output formats: html, xhtml, text, stext.

Keep imagess in output.
Do not expand ligature into constituent characters.
Do not merge spans on the same line.
Do not convert all whitespace into space characters.
Do not add spaces between gaps in the text.
Break blocks at paragraph boundaries.
Attempts to join up hyphenated words.
Do not apply ActualText replacements.
Use character code if unicode mapping fails.
Use glyph index if unicode mapping fails.
Calculate bboxes from the outlines.
Calculate ascender/descender from font glyphs.
Expand character bboxes to completely include width of glyphs.
Attempt to detect text features (fake bold, strikeout, underlined, etc.)
Do not include text that is completely clipped.
Specify clipping rectangle within which to collect content.
Collect structure markup.
Include vector bboxes in output.
Delay vectors in the extraction that would otherwise split a text line.
Merge abutting horizontal and vertical vectors.
Attempt to segment the page.
Hunt for tables within a (segmented) page.

PDF OUTPUT OPTIONS

These apply only to: PDF.

Decompress all streams (except compress-fonts/images).
Compress all streams, if no value is given, or the value is yes the default is to use flate.
Compress object stream for embedded fonts.
Compress object streams for images.
Effort spent compressing, 0 is default, otherwise percentage of max effort.
Encode object streams using ASCII hex encoding.
Pretty-print objects with indentation.
Print object labels.
No longer supported! Optimize document for progressive loading in viewers.
Pretty-print graphics commands in content streams.
Clean up graphics command in content streams.
Garbage collect unused objects.
Garbage collect unused object, and compact cross reference table.
Garbage collect unused objects, compact cross reference table, and recombine duplicate objects.
Write changes as incremental update.
Use object streams and cross reference streams.
Synthesize just missing, or all, annotation/widget appearance streams.
Continue saving the document even if there is an error.
This option is deprecated: instead of decrypt use encrypt=none.
Write encrypted document using specified encryption method. The default value is keep.
Document permissions to grant when encrypting.
Password required to read document.
Password required to edit document.
Regenerate document id (default yes).

EXIT STATUS

The exit status is 0 upon successful execution and non-zero upon error.

SEE ALSO

mupdf(1),

AUTHOR

MuPDF is Copyright 2004-2026 Artifex Software, Inc.

June 3, 2026