Reference, version 9.0.41

Contenta Converter CLI and MCP reference

contenta is the command line of Contenta Converter, installed with the desktop app. contenta serve runs its MCP server for AI agents. This page lists every command, option, exit code and MCP tool.

contenta <command> [arguments] [options]
contenta convert photo.cr2 -o out --format jpg

Generated from contenta catalog, version 9.0.41. Every command also has --help. Older spellings of some commands and options still work, so existing scripts keep running, but this page and --help leave them out. The same reference is on GitHub: cli.md and mcp-server.md.

Commands

Command	What it does
`contenta ai-transform <input>`	AI-powered image transformation via Google Gemini.
`contenta batch <input-dir>`	Batch convert images in a directory. With --json, stdout is one result object {"type":"summary","success":true\|false,"exitCode":N,"succeeded":N,...}. Add --progress for one line per finished output before it, {"type":"progress","file":NAME,"status":"done\|skipped\|failed","completed":N,"total":T} (the result object stays last): skip the lines that start with {"type":"progress".
`contenta catalog`	Print every CLI command and MCP tool as JSON (for AI agents, docs and the website).
`contenta convert <input>`	Convert a single image file.
`contenta effects`	List all available image effects.
`contenta formats`	List supported input and output formats.
`contenta info <file>`	Show format detection and metadata for an image file.
`contenta pdf-album <files>...`	Create a PDF photo album from multiple images. During the trial the PDF always carries the trial watermark.
`contenta pdf-merge <files>...`	Append multiple PDF files into a single PDF (page-level merge, no rasterization). During the trial the PDF always carries the trial watermark.
`contenta profile`	Save, load, or validate conversion profiles.
`contenta profile load <file>`	Load and display a profile from a JSON file.
`contenta profile save <file>`	Save current options as a profile JSON file.
`contenta profile validate <file>`	Validate a profile JSON file.
`contenta register <email> <key>`	Register your license key.
`contenta serve`	Start MCP server for AI agent integration.
`contenta slideshow <files>...`	Create a video slideshow from images. During the trial the video always ends with the branding end card.
`contenta status`	Show license status, tools availability, and version info.
`contenta watch <dir>`	Watch a directory and auto-convert new/changed images.

Output options

These options work the same way in the four ContentaSoft command lines. catalog and serve take none of them: catalog always prints JSON and serve keeps stdout for the MCP protocol.

Option	Description
`--json`	Print one JSON object on stdout (success, exitCode, error, warnings, errors, outputs + the command's fields); nothing else goes to stdout.
`--plain`	Plain text: no colours, no escape codes, ASCII tables. Automatic when stdout is a pipe or a file, or when NO_COLOR is set.
`--quiet`	Print errors only.
`--verbose`	Verbose output: details and, on an error, the stack trace.
`--progress`	With --json: before the result object, one line per finished item, {"type":"progress","file":NAME,"completed":N,"total":T}.

Exit codes

The same table in the four ContentaSoft command lines. --help ends with it and catalog lists it.

Code	Name	Meaning
0	`ok`	OK: everything asked for was done.
1	`unexpected_error`	Unexpected error (a bug or a broken install); see the log.
2	`invalid_arguments`	Invalid arguments: unknown option, missing or malformed value; nothing was done.
3	`input_not_found`	Input not found or unreadable; nothing was done.
4	`partial_failure`	Partial failure: some items succeeded, some failed.
5	`all_failed`	All failed: no item succeeded.
6	`license_refused`	License or trial refused the run.
130	`cancelled`	Cancelled (Ctrl+C).

JSON output

With --json a command prints one JSON object on stdout and nothing else; logs go to stderr. Every object has these keys, beside the fields of the command itself:

Key	Meaning
`success`	`true` when the run did everything it was asked to do
`exitCode`	the process exit code (see Exit codes)
`error`	one sentence when the run failed, otherwise `null`
`warnings`	notes about the run that did not stop it
`errors`	one entry per item that failed (a batch keeps going)
`outputs`	the files the run wrote

Trial keys, present only when they apply:

Key	Meaning
`trialWatermarked`	the output carries the trial watermark
`trialExport`	the output is a trial export
`trialNotice`	one sentence about the trial, for the user
`freeFilesRemaining`	free full-quality files left on this computer
`buyUrl`	where to buy a licence

With --json --progress a long command first prints one progress line per finished item (NDJSON), then the result object last. A script that wants only the result skips the lines that start with {"type":"progress". Progress keys:

Key	Meaning
`type`	always `"progress"`
`file`	the file that finished
`completed`	items finished so far
`total`	items in the run

Command reference

`contenta ai-transform`

AI-powered image transformation via Google Gemini.

contenta ai-transform <input> [options]

Argument	Description
`<input>`	Input image file path.

Option	Type	Description
`--prompt`	string	AI transformation prompt (required unless --remove-background).
`--api-key`	string	Gemini API key (or set GEMINI_API_KEY env var).
`--model`	string	Model tier: fast (default), pro or lite; or an explicit Gemini model ID.
`--size`	string	Output size: 1K (default), 2K or 4K (lite makes 1K only).
`--output`	string	Output file path.
`--remove-background`	boolean	Remove the background (no prompt needed; a given prompt is not used).
`--bg-mode`	string	Background mode: white, transparent, custom.
`--bg-color`	string	Custom background color (hex #RRGGBB).

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta batch`

Batch convert images in a directory. With --json, stdout is one result object {"type":"summary","success":true|false,"exitCode":N,"succeeded":N,...}. Add --progress for one line per finished output before it, {"type":"progress","file":NAME,"status":"done|skipped|failed","completed":N,"total":T} (the result object stays last): skip the lines that start with {"type":"progress".

contenta batch <input-dir> [options]

Argument	Description
`<input-dir>`	Input directory.

Option	Type	Description
`--include`	string[]	Only files matching this pattern (e.g. *.jpg, or jpg;png); repeat the option or separate patterns with ';' or ','.
`--exclude`	string[]	Leave out files matching this pattern: a glob such as .png or thumb_, or text in the path below the folder; repeat the option or separate patterns with ';' or ','.
`--recursive`	boolean	Include subdirectories.
`--workers`	integer	Parallel workers (default: automatic, see WorkerPolicy).
`--report`	string	Write a CSV report to this file: one row per output attempted, with the input, done/skipped/failed, the files written and the error.
`--output`	string	Output directory.
`--format`	string	Output format.
`--quality`	integer	Quality 1-100.
`--resize`	string	Resize to WxH.
`--resize-mode`	string	Resize mode: fit, fill, stretch, longest-edge, shortest-edge, fit-with-background, crop-to-aspect (default: fit). fit-with-background fits the picture inside --resize and fills the rest with --background.
`--background-color`	string	Colour around the picture with --resize-mode fit-with-background: white, black, grey, lightgrey or #RRGGBB (default: white). --resize 2000x2000 --resize-mode fit-with-background --background white makes white square product images.
`--resize-percent`	integer	Resize by percentage.
`--dont-enlarge`	boolean	Don't upscale smaller images.
`--dpi`	integer	Set DPI.
`--effects`	string[]	Effects, separated by spaces or commas, each name or name:key=value (e.g. sepia,sharpen:window=5). Run 'contenta effects' for names and parameters.
`--lut`	string	Apply a 3D LUT (.cube file) to every image, after resize and the other effects. With --profile it is added to the profile, in place of the profile's own LUT if it has one.
`--lut-strength`	integer	How much of the LUT to apply, 0 to 100 (default 100). Needs --lut, or a --profile that has a LUT.
`--watermark`	string	Add a white text watermark (width = 10% of the image).
`--watermark-position`	integer	Watermark position 0-8 (default: 8 = bottom-right).
`--watermark-opacity`	integer	Watermark opacity 0-100, 100 = fully visible (default: 70).
`--copyright`	string	Copyright metadata.
`--creator`	string	Creator metadata.
`--filename-pattern`	string	Output filename pattern (default: {originalname}). Tokens: {originalname} {counter:3} {today} {datetime} {size} and the Batch Rename tokens {name} {ext} {date} {date:FORMAT} {time} {year} {month} {camera} {make} {lens} {iso} {aperture} {focal} {w} {h} {gps} {seq} {seq:N} {folder}, filled from the source photo ({date} = date taken, {today} = the run date).
`--overwrite`	boolean	Overwrite existing.
`--skip-existing`	boolean	Skip an input whose output already exists (reported as skipped), so a re-run only converts what is missing.
`--color-space`	string	Color space: srgb, adobe-rgb, prophoto-rgb, preserve. srgb (the default) converts a source saved in another color space, such as an iPhone's Display P3; preserve leaves it as it is.
`--png-compression`	integer	PNG compression, zlib level 0-9 (0 = none/fastest, 9 = smallest; default 2).
`--webp-lossless`	boolean	WebP lossless.
`--jxl-lossless`	boolean	JXL lossless.
`--preserve-metadata`	boolean	Preserve metadata.
`--profile`	string	Load profile from JSON.
`--lens-corrections`	boolean	Apply the lens corrections a DNG carries (distortion, vignetting, shading; on by default; --lens-corrections false turns them off).
`--title`	string	Set title metadata.
`--description`	string	Set description metadata.
`--keywords`	string	Set keywords (comma-separated).
`--rights`	string	Set rights metadata.
`--creator-city`	string	Creator city.
`--creator-country`	string	Creator country.
`--creator-email`	string	Creator email.
`--creator-url`	string	Creator URL.
`--gps`	string	GPS coordinates (lat,lon or lat,lon,alt).
`--datetime`	string	Date/time override (ISO 8601).
`--aspect-ratio`	string	Custom aspect ratio (e.g., 16:9).
`--dpi-x`	integer	Horizontal DPI.
`--dpi-y`	integer	Vertical DPI.
`--sizes`	string	Emit multiple sizes at once: pixels (16,32,48,64) or scale factors (1x,2x,3x).
`--size-preset`	string	Built-in size set: favicon, appicon, srcset.
`--base-size`	integer	Base edge in px that scale factors (Nx) multiply (default: 512).
`--subfolder-mode`	string	Subfolder mode: flatten, preserve, template.
`--subfolder-pattern`	string	Subfolder template pattern.
`--backup-originals`	boolean	Backup originals.
`--backup-dir`	string	Backup directory.
`--pdf-page`	integer	Convert only this page (0-based: 0 is the first) of a multi-page PDF or TIFF. Without it every page is converted: one file each (name_page001.jpg, name_page002.jpg, ...), or one multi-page file when the output format is pdf or tiff.
`--zip-output`	boolean	After the batch, also pack every converted file into contenta-converter-images.zip in the output folder (the files stay in place; an existing archive of that name is replaced).
`--zip-split-size`	integer	With --zip-output: split into independent archives (contenta-converter-images_part2.zip, _part3...), each holding at most this many MB of files, every one opens on its own; a single larger file gets an archive to itself (default: 25).
`--rename-pattern`	string	After the batch, rename every converted file with the app's Batch Rename tokens: {name} {ext} {date} {date:FORMAT} {time} {year} {month} {camera} {make} {lens} {iso} {aperture} {focal} {w} {h} {gps} {seq} {seq:N} {folder}. The extension is kept; a name already taken gets _1, _2...
`--white-balance`	string	RAW white balance: camera, auto, manual.
`--color-temperature`	integer	Manual WB: temperature of the light the picture was taken in, in Kelvin (2000-10000, default 5500 = daylight; a lower value gives a cooler picture, a higher one a warmer picture).
`--denoising`	boolean	Apply wavelet denoising during RAW decode (off by default).
`--denoising-level`	integer	Denoising threshold (0-1000, default 300); giving a level turns denoising on.
`--sharpness-boost`	boolean	Sharpen after RAW decode (on by default for RAW input, as in the app; --sharpness-boost false turns it off; off by default with --raw-16bit).
`--color-boost`	boolean	Apply saturation boost after RAW decode.
`--contrast-boost`	boolean	Apply contrast boost after RAW decode.
`--exposure`	number	RAW exposure compensation in stops (EV), -3.0 to 3.0, default 0: +1 doubles the light, -1 halves it.
`--fast-raw-decode`	boolean	Faster RAW decode at half resolution when the output is at most half the size of the sensor image (slightly softer; for web and preview exports).
`--raw-16bit`	boolean	Decode RAW at 16 bits per channel and write a 16-bit TIFF. Needs -f tiff and no effects, RAW boosts, watermark or embedded preview (they work on 8 bits per channel); otherwise the file is written with 8 bits and the result carries a warning. Turns the default RAW sharpening off.
`--embedded-preview`	boolean	Convert the JPEG preview the camera stored in the RAW file instead of developing the RAW data: the fastest way, but the preview can be smaller than the full picture, already has the camera's look, and is 8-bit; white balance, denoising and exposure do not apply to it. A file without a preview is developed as usual.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta catalog`

Print every CLI command and MCP tool as JSON (for AI agents, docs and the website).

contenta catalog

`contenta convert`

Convert a single image file.

contenta convert <input> [options]

Argument	Description
`<input>`	Input image file path.

Option	Type	Description
`--output`	string	Output directory (default: same as input).
`--format`	string	Output format (jpg, png, webp, tiff, bmp, gif, jxl, heic, avif, svg, pdf, ico).
`--quality`	integer	Quality 1-100.
`--resize`	string	Resize to WxH (e.g., 1920x1080).
`--resize-mode`	string	Resize mode: fit, fill, stretch, longest-edge, shortest-edge, fit-with-background, crop-to-aspect (default: fit). fit-with-background fits the picture inside --resize and fills the rest with --background.
`--background-color`	string	Colour around the picture with --resize-mode fit-with-background: white, black, grey, lightgrey or #RRGGBB (default: white). --resize 2000x2000 --resize-mode fit-with-background --background white makes white square product images.
`--resize-percent`	integer	Resize by percentage (1-100).
`--dont-enlarge`	boolean	Don't upscale smaller images.
`--dpi`	integer	Set output DPI.
`--effects`	string[]	Effects, separated by spaces or commas, each name or name:key=value (e.g. sepia,sharpen:window=5). Run 'contenta effects' for names and parameters.
`--lut`	string	Apply a 3D LUT (.cube file) to every image, after resize and the other effects. With --profile it is added to the profile, in place of the profile's own LUT if it has one.
`--lut-strength`	integer	How much of the LUT to apply, 0 to 100 (default 100). Needs --lut, or a --profile that has a LUT.
`--watermark`	string	Add a white text watermark (width = 10% of the image).
`--watermark-position`	integer	Watermark position 0-8 (default: 8 = bottom-right).
`--watermark-opacity`	integer	Watermark opacity 0-100, 100 = fully visible (default: 70).
`--copyright`	string	Set copyright metadata.
`--creator`	string	Set creator metadata.
`--filename-pattern`	string	Output filename pattern (default: {originalname}). Tokens: {originalname} {counter:3} {today} {datetime} {size} and the Batch Rename tokens {name} {ext} {date} {date:FORMAT} {time} {year} {month} {camera} {make} {lens} {iso} {aperture} {focal} {w} {h} {gps} {seq} {seq:N} {folder}, filled from the source photo ({date} = date taken, {today} = the run date).
`--overwrite`	boolean	Overwrite existing files.
`--skip-existing`	boolean	Skip an input whose output already exists (reported as skipped), so a re-run only converts what is missing.
`--color-space`	string	Color space: srgb, adobe-rgb, prophoto-rgb, preserve. srgb (the default) converts a source saved in another color space, such as an iPhone's Display P3; preserve leaves it as it is.
`--png-compression`	integer	PNG compression, zlib level 0-9 (0 = none/fastest, 9 = smallest; default 2).
`--webp-lossless`	boolean	WebP lossless mode.
`--jxl-lossless`	boolean	JXL lossless mode.
`--preserve-metadata`	boolean	Preserve original metadata.
`--profile`	string	Load conversion profile from JSON file.
`--lens-corrections`	boolean	Apply the lens corrections a DNG carries (distortion, vignetting, shading; on by default; --lens-corrections false turns them off).
`--title`	string	Set title metadata.
`--description`	string	Set description metadata.
`--keywords`	string	Set keywords metadata (comma-separated).
`--rights`	string	Set rights/usage terms metadata.
`--creator-city`	string	Set creator city metadata.
`--creator-country`	string	Set creator country metadata.
`--creator-email`	string	Set creator email metadata.
`--creator-url`	string	Set creator URL metadata.
`--gps`	string	Set GPS coordinates (lat,lon or lat,lon,alt).
`--datetime`	string	Set date/time override (ISO 8601).
`--aspect-ratio`	string	Custom aspect ratio (e.g., 16:9).
`--dpi-x`	integer	Set horizontal DPI.
`--dpi-y`	integer	Set vertical DPI.
`--sizes`	string	Emit multiple sizes at once: pixels (16,32,48,64) or scale factors (1x,2x,3x).
`--size-preset`	string	Built-in size set: favicon, appicon, srcset.
`--base-size`	integer	Base edge in px that scale factors (Nx) multiply (default: 512).
`--subfolder-mode`	string	Subfolder mode: flatten, preserve, template.
`--subfolder-pattern`	string	Subfolder template pattern (e.g., {year}/{month}).
`--backup-originals`	boolean	Backup original files before conversion.
`--backup-dir`	string	Backup directory path.
`--pdf-page`	integer	Convert only this page (0-based: 0 is the first) of a multi-page PDF or TIFF. Without it every page is converted: one file each (name_page001.jpg, name_page002.jpg, ...), or one multi-page file when the output format is pdf or tiff.
`--white-balance`	string	RAW white balance: camera, auto, manual.
`--color-temperature`	integer	Manual WB: temperature of the light the picture was taken in, in Kelvin (2000-10000, default 5500 = daylight; a lower value gives a cooler picture, a higher one a warmer picture).
`--denoising`	boolean	Apply wavelet denoising during RAW decode (off by default).
`--denoising-level`	integer	Denoising threshold (0-1000, default 300); giving a level turns denoising on.
`--sharpness-boost`	boolean	Sharpen after RAW decode (on by default for RAW input, as in the app; --sharpness-boost false turns it off; off by default with --raw-16bit).
`--color-boost`	boolean	Apply saturation boost after RAW decode.
`--contrast-boost`	boolean	Apply contrast boost after RAW decode.
`--exposure`	number	RAW exposure compensation in stops (EV), -3.0 to 3.0, default 0: +1 doubles the light, -1 halves it.
`--fast-raw-decode`	boolean	Faster RAW decode at half resolution when the output is at most half the size of the sensor image (slightly softer; for web and preview exports).
`--raw-16bit`	boolean	Decode RAW at 16 bits per channel and write a 16-bit TIFF. Needs -f tiff and no effects, RAW boosts, watermark or embedded preview (they work on 8 bits per channel); otherwise the file is written with 8 bits and the result carries a warning. Turns the default RAW sharpening off.
`--embedded-preview`	boolean	Convert the JPEG preview the camera stored in the RAW file instead of developing the RAW data: the fastest way, but the preview can be smaller than the full picture, already has the camera's look, and is 8-bit; white balance, denoising and exposure do not apply to it. A file without a preview is developed as usual.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta effects`

List all available image effects.

contenta effects

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta formats`

List supported input and output formats.

contenta formats [options]

Option	Type	Description
`--input`	boolean	Show input formats only.
`--output`	boolean	Show output formats only.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta info`

Show format detection and metadata for an image file.

contenta info <file>

Argument	Description
`<file>`	Image file path.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta pdf-album`

Create a PDF photo album from multiple images. During the trial the PDF always carries the trial watermark.

contenta pdf-album <files>... [options]

Argument	Description
`<files>`	Input image files.

Option	Type	Description
`--output`	string	Output PDF path.
`--photos-per-page`	integer	Photos per page (1, 2, 4, 6, 8, 10, 12, 16, 24, 48).
`--page-size`	string	Page size: A4, Letter, Legal.
`--orientation`	string	Orientation: Auto, Portrait, Landscape.
`--background`	string	Background color (hex, e.g., #FF000000).
`--background-image`	string	Background image path.
`--background-mode`	string	Background image mode: stretch, center, tile.
`--title-page`	string	Title page image path.
`--title-page-mode`	string	Title page image mode: stretch, center, tile.
`--show-filenames`	boolean	Show filenames below photos.
`--show-filenames-color`	string	Filename label color (hex, e.g., #FFFFFFFF).
`--embed-quality`	integer	JPEG quality 1-100 for non-JPEG inputs (PNG, TIFF, RAW...); JPEG inputs are embedded unchanged.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta pdf-merge`

Append multiple PDF files into a single PDF (page-level merge, no rasterization). During the trial the PDF always carries the trial watermark.

contenta pdf-merge <files>... [options]

Argument	Description
`<files>`	Input PDF files (appended in argument order).

Option	Type	Description
`--output`	string	Output PDF path.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta profile`

Save, load, or validate conversion profiles.

Subcommands: profile load, profile save, profile validate.

`contenta profile load`

Load and display a profile from a JSON file.

contenta profile load <file>

Argument	Description
`<file>`	Profile JSON path to load.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta profile save`

Save current options as a profile JSON file.

contenta profile save <file> [options]

Argument	Description
`<file>`	Output profile JSON path.

Option	Type	Description
`--format`	string	Output format.
`--quality`	integer	Quality 1-100.
`--resize`	string	Resize to WxH.
`--resize-mode`	string	Resize mode: fit, fill, stretch, longest-edge, shortest-edge, fit-with-background, crop-to-aspect (default: fit). fit-with-background fits the picture inside --resize and fills the rest with --background.
`--background-color`	string	Colour around the picture with --resize-mode fit-with-background: white, black, grey, lightgrey or #RRGGBB (default: white). --resize 2000x2000 --resize-mode fit-with-background --background white makes white square product images.
`--effects`	string[]	Effects, separated by spaces or commas, each name or name:key=value (e.g. sepia,sharpen:window=5). Run 'contenta effects' for names and parameters.
`--copyright`	string	Copyright metadata.
`--creator`	string	Creator metadata.
`--overwrite`	boolean	Overwrite existing files.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta profile validate`

Validate a profile JSON file.

contenta profile validate <file>

Argument	Description
`<file>`	Profile JSON path to validate.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta register`

contenta register <email> <key>

Argument	Description
`<email>`	Registration email.
`<key>`	25-character license key.

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta serve`

Start MCP server for AI agent integration.

contenta serve

`contenta slideshow`

Create a video slideshow from images. During the trial the video always ends with the branding end card.

contenta slideshow <files>... [options]

Argument	Description
`<files>`	Input image files.

Option	Type	Description
`--output`	string	Output video file path.
`--template`	string	Template: tiktok, shorts, facebook-reels, snapchat, youtube, linkedin, twitter, instagram, pinterest.
`--duration`	integer	Duration per slide in ms.
`--transition`	integer	Transition duration in ms.
`--ken-burns`	string	Ken Burns mode: zoom-in, zoom-out, pan-left, pan-right, pan-up, pan-down, alternating, off.
`--no-branding`	boolean	Disable branding end card (registered copies only; during the trial the end card is always added).
`--audio`	string	Background audio file path (MP3, WAV, M4A).
`--loop-audio`	boolean	Loop audio if shorter than video.
`--format`	string	Output format: mp4, webm.
`--quality`	integer	CRF quality (0-51, lower = better).
`--fps`	integer	Frames per second.
`--encoder`	string	Video encoder: auto, libx264, h264_nvenc, h264_qsv, h264_amf.
`--text-overlay`	string[]	Text overlay (format: text or slide:text:position).

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta status`

Show license status, tools availability, and version info.

contenta status

Output options: --json, --plain, --quiet, --verbose, --progress.

`contenta watch`

Watch a directory and auto-convert new/changed images.

contenta watch <dir> [options]

Argument	Description
`<dir>`	Directory to watch.

Option	Type	Description
`--include`	string[]	Only files matching this pattern (e.g. *.jpg, or jpg;png); repeat the option or separate patterns with ';' or ','.
`--exclude`	string[]	Leave out files matching this pattern (as batch --exclude: a glob such as .png or thumb_, or text in the path below the folder); repeat the option or separate patterns with ';' or ','.
`--debounce`	integer	Debounce delay in ms.
`--recursive`	boolean	Watch subdirectories.
`--output`	string	Output directory.
`--format`	string	Output format.
`--quality`	integer	Quality 1-100.
`--profile`	string	Load profile from JSON.
`--white-balance`	string	RAW white balance: camera, auto, manual.
`--color-temperature`	integer	Manual WB: temperature of the light the picture was taken in, in Kelvin (2000-10000, default 5500 = daylight; a lower value gives a cooler picture, a higher one a warmer picture).
`--denoising`	boolean	Apply wavelet denoising during RAW decode (off by default).
`--denoising-level`	integer	Denoising threshold (0-1000, default 300); giving a level turns denoising on.
`--sharpness-boost`	boolean	Sharpen after RAW decode (on by default for RAW input, as in the app; --sharpness-boost false turns it off; off by default with --raw-16bit).
`--color-boost`	boolean	Apply saturation boost after RAW decode.
`--contrast-boost`	boolean	Apply contrast boost after RAW decode.
`--exposure`	number	RAW exposure compensation in stops (EV), -3.0 to 3.0, default 0: +1 doubles the light, -1 halves it.
`--fast-raw-decode`	boolean	Faster RAW decode at half resolution when the output is at most half the size of the sensor image (slightly softer; for web and preview exports).
`--raw-16bit`	boolean	Decode RAW at 16 bits per channel and write a 16-bit TIFF. Needs -f tiff and no effects, RAW boosts, watermark or embedded preview (they work on 8 bits per channel); otherwise the file is written with 8 bits and the result carries a warning. Turns the default RAW sharpening off.
`--embedded-preview`	boolean	Convert the JPEG preview the camera stored in the RAW file instead of developing the RAW data: the fastest way, but the preview can be smaller than the full picture, already has the camera's look, and is 8-bit; white balance, denoising and exposure do not apply to it. A file without a preview is developed as usual.

Output options: --json, --plain, --quiet, --verbose, --progress.

Effects

Effect names for --effects and the MCP effects parameter, as contenta effects lists them.

Effect	Category	What it does	Parameters
`blackwhite`	Color	Black & White	r: 0 to 100 (default 30), g: 0 to 100 (default 59), b: 0 to 100 (default 11)
`sepia`	Color	Sepia	none
`colorbrightnesscontrast`	Color	Brightness / Contrast	brightness: -100 to 100 (default 0), contrast: -100 to 100 (default 0), gamma: 0.1 to 5.0 (default 1.0), invert: true\|false
`lighten`	Color	Lighten	level: 1 to 100 (default 30)
`darken`	Color	Darken	level: 1 to 100 (default 30)
`contrast`	Color	Contrast Only (maps to fastc 'colors')	level: -100 to 100 (default 20)
`colortone`	Color	Color Tone	red: -100 to 100 (default 0), green: -100 to 100 (default 0), blue: -100 to 100 (default 0), saturation: 0 to 100 (default 50)
`huesaturation`	Color	Hue / Saturation	hue: -180 to 180 (default 0), saturation: -100 to 100 (default 0)
`autocolorlevel`	Color	Auto Color Level	none
`colorlevel`	Color	Color Levels	input_low: 0 to 255 (default 0), input_high: 0 to 255 (default 255), output_low: 0 to 255 (default 0), output_high: 0 to 255 (default 255), gamma: 0.1 to 5.0 (default 1.0), color_channel: 0 to 3 (default 0)
`colorbalance`	Color	Color Balance	preserve_lum: true\|false, tone_region: 0 to 2 (default 1), cyan_red: -100 to 100 (default 0), magenta_green: -100 to 100 (default 0), yellow_blue: -100 to 100 (default 0)
`colortemp`	Color	Color Temperature	temperature: -100 to 100 (default 0), 0 = unchanged, negative = cooler, positive = warmer
`invert`	Color	Invert Colors	none
`histogramequalization`	Color	Histogram Equalization	none
`saturationequalization`	Color	Saturation Equalization	none
`splash`	Color	Splash (Color Keep)	level: 0 to 100 (default 50)
`posterize`	Color	Posterize	level: 2 to 32 (default 6)
`threshold`	Color	Threshold	level: 0 to 255 (default 128)
`solarize`	Color	Solarize	level: 0 to 255 (default 128)
`applylut`	Color	Apply a 3D LUT (.cube, in-process)	path: a .cube file, strength: 0 to 100 (default 100)
`autoenhance`	Enhance	Auto-Enhance (in-process)	strength: 0 to 100 (default 50)
`clarity`	Enhance	Clarity (in-process)	amount: 0 to 100 (default 40)
`dehaze`	Enhance	Dehaze (in-process)	amount: 0 to 100 (default 50)
`aidenoise`	Enhance	AI Denoise (in-process)	strength: 0 to 100 (default 50)
`blur`	Blur / Sharpen	Blur	window: 1 to 30 (default 3)
`sharpen`	Blur / Sharpen	Sharpen	window: 1 to 30 (default 3)
`median`	Blur / Sharpen	Median Filter	window: 1 to 10 (default 2)
`refocus`	Blur / Sharpen	Refocus (Unsharp Mask)	window: 1 to 20 (default 5)
`motionblur`	Blur / Sharpen	Motion Blur	speed: 1 to 50 (default 10), angle: 0 to 360 (default 0)
`radialblur`	Blur / Sharpen	Radial Blur	level: 1 to 30 (default 5)
`zoomblur`	Blur / Sharpen	Zoom Blur	level: 1 to 30 (default 5)
`pencilsketch`	Artistic	Pencil Sketch	pensize: 1 to 20 (default 3), rough: 0 to 100 (default 50)
`oilpaint`	Artistic	Oil Paint	pensize: 1 to 20 (default 4), rough: 0 to 100 (default 50)
`emboss`	Artistic	Emboss	angle: 0 to 360 (default 45)
`oldphoto`	Artistic	Old Photo	window: 1 to 20 (default 3)
`glow`	Artistic	Glow	soft: 1 to 30 (default 5), brightness: -50 to 50 (default 10), contrast: -50 to 50 (default -10)
`softportrait`	Artistic	Soft Portrait	soft: 1 to 30 (default 5), brightness: -50 to 50 (default 5), warm: 0 to 30 (default 10)
`softedge`	Artistic	Soft Edge	soft: 1 to 30 (default 5)
`noisy`	Artistic	Add Noise	amount: 1 to 50 (default 10), iscolor: true\|false
`mosaic`	Distortion	Mosaic / Pixelate	size: 2 to 80 (default 10)
`shift`	Distortion	Shift	level: 1 to 50 (default 10)
`ribbon`	Distortion	Ribbon	amplitude: 1 to 30 (default 10), frequency: 1 to 20 (default 5)
`bulge`	Distortion	Bulge	level: -100 to 100 (default 40)
`twist`	Distortion	Twist	twist: -90 to 90 (default 30), size: 10 to 100 (default 50)
`illusion`	Distortion	Illusion	level: 1 to 20 (default 5)
`wave`	Distortion	Wave	wavelength: 10 to 200 (default 40), amplitude: 1 to 50 (default 10)
`ripple`	Distortion	Ripple	length: 10 to 200 (default 40), amount: 1 to 50 (default 10), type: sin\|square
`perspective`	Correction	Perspective Correction (in-process)	vertical: -100 to 100 (default 0), horizontal: -100 to 100 (default 0)
`crop`	Transforms	Crop image (in-process)	left: 0 to 100, top: 0 to 100, right: 0 to 100, bottom: 0 to 100 (percent of the picture trimmed from each edge, default 0); or x, y, width, height: pixels from the top-left corner. The crop cuts the picture first and a resize then sizes what is left: both forms count on the picture BEFORE the resize (after a flip or rotate placed before the crop), so the rectangle is in pixels of the source, not of the resized output
`flip`	Transforms	Flip (in-process)	direction: horizontal\|vertical (default horizontal)
`rotate`	Transforms	Rotate (in-process)	angle: 0 to 360 (default 90)
`watermark`	Transforms	Text/image watermark (in-process)	text, placement 0-8, transparency 0-100, font

MCP server

contenta serve exposes 13 tools over the Model Context Protocol (stdio, newline-delimited JSON-RPC 2.0; protocol versions 2025-11-25, 2025-06-18, 2025-03-26 and 2024-11-05), so an AI agent such as Claude Code, Claude Desktop, Cursor or Codex can work on files on your PC. Server name: contenta-converter. Pass absolute paths to every tool.

{
  "mcpServers": {
    "contenta-converter": {
      "command": "contenta",
      "args": [
        "serve"
      ]
    }
  }
}

In Claude Code one command adds it: claude mcp add contenta-converter -- contenta serve. Stdout carries JSON-RPC only, ping is answered while a tool runs, notifications/cancelled stops a running call, and a tool that cannot do its job returns isError: true with a sentence that names the problem.

Tool	Title	Read-only	Destructive	Idempotent	Network
`ai_transform`	AI transform image (Gemini)	no	yes	no	yes
`batch_convert`	Batch convert images	no	no	no	no
`convert_image`	Convert image	no	no	no	no
`create_pdf_album`	Create PDF album	no	yes	no	no
`create_slideshow`	Create video slideshow	no	yes	no	no
`detect_format`	Detect image format	yes	no	yes	no
`get_file_info`	Get image file info	yes	no	yes	no
`get_license_status`	Get license and trial status	yes	no	yes	no
`list_effects`	List effects	yes	no	yes	no
`list_formats`	List supported formats	yes	no	yes	no
`merge_pdfs`	Merge PDFs	no	yes	no	no
`read_metadata`	Read image metadata	yes	no	yes	no
`write_metadata`	Write image metadata	no	yes	no	no

Destructive: the tool can replace a file that already exists. Network: the tool uses the internet.

`ai_transform`

writes filescan replace filesuses the internet

AI transform image (Gemini). AI-powered image transformation via Google Gemini (edit, remove background, enhance). During the trial, results beyond the free clean outputs carry the trial watermark.

Parameter	Type	Required	Description
`input_path`	string	yes	Absolute path to the source image.
`prompt`	string	no	Transformation prompt describing what to do (e.g., 'Make it brighter'). Required unless remove_background is true.
`api_key`	string	no	Google Gemini API key (or set GEMINI_API_KEY env var).
`output_path`	string	no	Output file path (default: {name}_ai.{ext} next to input).
`model`	string	no	Model tier: fast (default), pro or lite; or an explicit Gemini model ID.
`size`	string	no	Output size: 1K (default), 2K or 4K (lite makes 1K only). One of: `1K`, `2K`, `4K`.
`transparent_background`	boolean	no	Request transparent background (PNG output).
`remove_background`	boolean	no	Remove the background; no prompt needed (the CLI's --remove-background).
`bg_mode`	string	no	With remove_background: white (default), transparent or custom. One of: `white`, `transparent`, `custom`.
`bg_color`	string	no	With bg_mode custom: the background colour as #RRGGBB.

`batch_convert`

writes files

Batch convert images. Convert multiple images at once. Provide file paths or a directory. A multi-page PDF or TIFF converts every page: one output file per page, or one multi-page file when the output format is PDF or TIFF (pdf_page picks a single page); total counts outputs, inputs counts the files given.

Parameter	Type	Required	Description
`input_paths`	array of string	no	Array of absolute file paths to convert.
`input_dir`	string	no	Directory to scan for images (alternative to input_paths).
`output_dir`	string	no	Output directory.
`overwrite`	boolean	no	Replace an output file that already exists (default: false, a new name is chosen instead).
`skip_existing`	boolean	no	Leave an input alone when its output already exists (reported as skipped), so a re-run only converts what is missing (default: false).
`format`	string	no	Output format (jpg, png, webp, etc.)
`quality`	integer	no	Output quality. Range: 1 to 100.
`resize_width`	integer	no	Target width (alone: the height follows the aspect ratio). Minimum: 1.
`resize_height`	integer	no	Target height (alone: the width follows the aspect ratio). Minimum: 1.
`resize_mode`	string	no	Resize mode: fit (default), fill, stretch, longest-edge, shortest-edge, fit-with-background, crop-to-aspect. fill, stretch, fit-with-background and crop-to-aspect need both resize_width and resize_height.
`background`	string	no	Colour around the picture with resize_mode fit-with-background: white, black, grey, lightgrey or #RRGGBB (default: white). resize_width 2000 + resize_height 2000 + resize_mode fit-with-background + background white = white square product images.
`effects`	array of string	no	Effects to apply in order, each name or name:key=value (e.g. ["sepia", "sharpen:window=5"]; a comma list "sepia,sharpen" works too). Names and parameters: list_effects. A crop cuts the picture BEFORE the resize (its x, y, width, height are pixels of the source; left, top, right, bottom are percent): the resize then sizes what is left. An unknown name fails the call before anything is converted.
`lut_path`	string	no	Absolute path to a 3D LUT (.cube) applied to every image, after resize and the other effects (the CLI's --lut). A file that is missing or not a 3D .cube LUT fails the call before anything is converted.
`lut_strength`	integer	no	How much of the LUT to apply, 0-100 (default 100). Needs lut_path. Range: 0 to 100.
`watermark_text`	string	no	Text watermark drawn on every output in white, its width 10% of the image (the CLI's --watermark). Nothing is drawn without it.
`watermark_position`	integer	no	Where the watermark goes, 0-8 read left to right, top to bottom: 0 top-left, 1 top-centre, 2 top-right, 3 middle-left, 4 centre, 5 middle-right, 6 bottom-left, 7 bottom-centre, 8 bottom-right (default: 8). Range: 0 to 8.
`watermark_opacity`	integer	no	Watermark opacity 0-100, 100 = fully visible (default: 70). Range: 0 to 100.
`lens_corrections`	boolean	no	Apply the lens corrections a DNG file carries: distortion, vignetting, shading (default: true). No effect on other inputs.
`pdf_page`	integer	no	Multi-page PDF or TIFF: convert only this page (0-based: 0 is the first page). Omit it to convert EVERY page: one output file each (name_page001.jpg, name_page002.jpg, ...), or one multi-page file when the output format is pdf or tiff; the result lists every file in outputs. Minimum: 0.
`white_balance`	string	no	RAW input only. RAW white balance: camera, auto, manual (default: camera). One of: `camera`, `auto`, `manual`.
`color_temperature`	integer	no	RAW input, white_balance manual. Manual WB: temperature of the light the picture was taken in, in Kelvin (2000-10000, default 5500 = daylight; a lower value gives a cooler picture, a higher one a warmer picture). Range: 2000 to 10000.
`denoising`	boolean	no	RAW input only. Apply wavelet denoising during RAW decode (off by default).
`denoising_level`	integer	no	RAW input only. Denoising threshold (0-1000, default 300); giving a level turns denoising on. Range: 0 to 1000.
`sharpness_boost`	boolean	no	RAW input only. Sharpen after RAW decode (on by default for RAW input, as in the app; false turns it off; off by default with raw_16bit).
`color_boost`	boolean	no	RAW input only. Apply saturation boost after RAW decode.
`contrast_boost`	boolean	no	RAW input only. Apply contrast boost after RAW decode.
`exposure`	number	no	RAW input only. RAW exposure compensation in stops (EV), -3.0 to 3.0, default 0: +1 doubles the light, -1 halves it. Range: -3 to 3.
`fast_raw_decode`	boolean	no	RAW input only. Faster RAW decode at half resolution when the output is at most half the size of the sensor image (slightly softer; for web and preview exports).
`raw_16bit`	boolean	no	RAW input only. Decode RAW at 16 bits per channel and write a 16-bit TIFF. Needs format tiff and no effects, RAW boosts, watermark or embedded preview (they work on 8 bits per channel); otherwise the file is written with 8 bits and the result carries a warning. Turns the default RAW sharpening off; a file written with 8 bits instead is named in the result's warnings.
`embedded_preview`	boolean	no	RAW input only. Convert the JPEG preview the camera stored in the RAW file instead of developing the RAW data: the fastest way, but the preview can be smaller than the full picture, already has the camera's look, and is 8-bit; white balance, denoising and exposure do not apply to it. A file without a preview is developed as usual (default: false).
`workers`	integer	no	Parallel workers (default: automatic - one per logical CPU, 2x for JPEG downscales to half size or less, one per physical core for RAW batches on 8+ core CPUs). Minimum: 1.

`convert_image`

writes files

Convert image. Convert a single image to a different format with optional resize, effects, and metadata. A multi-page PDF or TIFF converts every page: one output file per page, or one multi-page file when the output format is PDF or TIFF (pdf_page picks a single page); the result lists every file in outputs.

Parameter	Type	Required	Description
`input_path`	string	yes	Absolute path to the source image.
`output_dir`	string	no	Output directory (default: same as input).
`overwrite`	boolean	no	Replace an output file that already exists (default: false, a new name is chosen instead).
`skip_existing`	boolean	no	Leave an input alone when its output already exists (reported as skipped), so a re-run only converts what is missing (default: false).
`format`	string	no	Output format: jpg, png, webp, tiff, bmp, gif, jxl, heic, avif, svg, pdf.
`quality`	integer	no	Output quality (default: 90). Range: 1 to 100.
`resize_width`	integer	no	Target width in pixels (alone: the height follows the aspect ratio). Minimum: 1.
`resize_height`	integer	no	Target height in pixels (alone: the width follows the aspect ratio). Minimum: 1.
`resize_mode`	string	no	Resize mode: fit (default), fill, stretch, longest-edge, shortest-edge, fit-with-background, crop-to-aspect. fill, stretch, fit-with-background and crop-to-aspect need both resize_width and resize_height.
`background`	string	no	Colour around the picture with resize_mode fit-with-background: white, black, grey, lightgrey or #RRGGBB (default: white). resize_width 2000 + resize_height 2000 + resize_mode fit-with-background + background white = white square product images.
`effects`	array of string	no	Effects to apply in order, each name or name:key=value (e.g. ["sepia", "sharpen:window=5"]; a comma list "sepia,sharpen" works too). Names and parameters: list_effects. A crop cuts the picture BEFORE the resize (its x, y, width, height are pixels of the source; left, top, right, bottom are percent): the resize then sizes what is left. An unknown name fails the call before anything is converted.
`lut_path`	string	no	Absolute path to a 3D LUT (.cube) applied to every image, after resize and the other effects (the CLI's --lut). A file that is missing or not a 3D .cube LUT fails the call before anything is converted.
`lut_strength`	integer	no	How much of the LUT to apply, 0-100 (default 100). Needs lut_path. Range: 0 to 100.
`watermark_text`	string	no	Text watermark drawn on every output in white, its width 10% of the image (the CLI's --watermark). Nothing is drawn without it.
`watermark_position`	integer	no	Where the watermark goes, 0-8 read left to right, top to bottom: 0 top-left, 1 top-centre, 2 top-right, 3 middle-left, 4 centre, 5 middle-right, 6 bottom-left, 7 bottom-centre, 8 bottom-right (default: 8). Range: 0 to 8.
`watermark_opacity`	integer	no	Watermark opacity 0-100, 100 = fully visible (default: 70). Range: 0 to 100.
`copyright`	string	no	Copyright metadata to embed.
`creator`	string	no	Creator metadata to embed.
`preserve_metadata`	boolean	no	Preserve original metadata (default: true).
`white_balance`	string	no	RAW input only. RAW white balance: camera, auto, manual (default: camera). One of: `camera`, `auto`, `manual`.
`color_temperature`	integer	no	RAW input, white_balance manual. Manual WB: temperature of the light the picture was taken in, in Kelvin (2000-10000, default 5500 = daylight; a lower value gives a cooler picture, a higher one a warmer picture). Range: 2000 to 10000.
`denoising`	boolean	no	RAW input only. Apply wavelet denoising during RAW decode (off by default).
`denoising_level`	integer	no	RAW input only. Denoising threshold (0-1000, default 300); giving a level turns denoising on. Range: 0 to 1000.
`sharpness_boost`	boolean	no	RAW input only. Sharpen after RAW decode (on by default for RAW input, as in the app; false turns it off; off by default with raw_16bit).
`color_boost`	boolean	no	RAW input only. Apply saturation boost after RAW decode.
`contrast_boost`	boolean	no	RAW input only. Apply contrast boost after RAW decode.
`exposure`	number	no	RAW input only. RAW exposure compensation in stops (EV), -3.0 to 3.0, default 0: +1 doubles the light, -1 halves it. Range: -3 to 3.
`fast_raw_decode`	boolean	no	RAW input only. Faster RAW decode at half resolution when the output is at most half the size of the sensor image (slightly softer; for web and preview exports).
`raw_16bit`	boolean	no	RAW input only. Decode RAW at 16 bits per channel and write a 16-bit TIFF. Needs format tiff and no effects, RAW boosts, watermark or embedded preview (they work on 8 bits per channel); otherwise the file is written with 8 bits and the result carries a warning. Turns the default RAW sharpening off; a file written with 8 bits instead is named in the result's warnings.
`embedded_preview`	boolean	no	RAW input only. Convert the JPEG preview the camera stored in the RAW file instead of developing the RAW data: the fastest way, but the preview can be smaller than the full picture, already has the camera's look, and is 8-bit; white balance, denoising and exposure do not apply to it. A file without a preview is developed as usual (default: false).
`lens_corrections`	boolean	no	Apply the lens corrections a DNG file carries: distortion, vignetting, shading (default: true). No effect on other inputs.
`pdf_page`	integer	no	Multi-page PDF or TIFF: convert only this page (0-based: 0 is the first page). Omit it to convert EVERY page: one output file each (name_page001.jpg, name_page002.jpg, ...), or one multi-page file when the output format is pdf or tiff; the result lists every file in outputs. Minimum: 0.

`create_pdf_album`

writes filescan replace files

Create PDF album. Create a PDF photo album from multiple images with configurable layout. During the trial the PDF always carries the trial watermark.

Parameter	Type	Required	Description
`input_paths`	array of string	yes	Required. Absolute paths to images.
`output_path`	string	yes	Output PDF file path.
`photos_per_page`	integer	no	Photos per page: 1, 2, 4, 6, 8, 10, 12, 16, 24, 48 (default: 4).
`page_size`	string	no	Page size: A4, Letter, Legal (default: A4).
`orientation`	string	no	Orientation: Auto, Portrait, Landscape (default: Auto).
`embed_quality`	integer	no	JPEG quality for non-JPEG inputs; JPEG inputs are embedded unchanged (default: 85). Range: 1 to 100.

`create_slideshow`

writes filescan replace files

Create video slideshow. Create a video slideshow from multiple images with Ken Burns motion, transitions, and optional audio. During the trial the video always ends with the branding end card.

Parameter	Type	Required	Description
`input_paths`	array of string	yes	Required. Array of image file paths (minimum 2).
`output_path`	string	yes	Output video file path.
`template`	string	no	Video template: tiktok (9:16), shorts (9:16), facebook-reels (9:16), snapchat (9:16), youtube (16:9), linkedin (16:9), twitter (16:9), instagram (1:1), pinterest (2:3). Default: youtube.
`duration_ms`	integer	no	Duration per slide in milliseconds (default: 3000).
`transition_ms`	integer	no	Transition duration in milliseconds (default: 800).
`ken_burns`	string	no	Ken Burns mode: zoom-in, zoom-out, pan-left, pan-right, alternating, off. Default: zoom-in.
`format`	string	no	Output format: mp4, webm (default: mp4).
`quality`	integer	no	CRF quality (default: 23, lower = better). Range: 0 to 51.
`audio_path`	string	no	Background audio file path.
`add_branding`	boolean	no	Add branding end card (default: true). Registered copies only: during the trial the end card is always added and false is ignored.

`detect_format`

read-onlyidempotent

Detect image format. Detect the image format of a file (RAW, HEIC, JXL, PSD, PDF, etc.)

Parameter	Type	Required	Description
`input_path`	string	yes	Required. Absolute path to the image file.

`get_file_info`

read-onlyidempotent

Get image file info. Inspect an image file (the CLI's contenta info): real format, pixel size, DPI, file size, copyright and creator.

Parameter	Type	Required	Description
`input_path`	string	yes	Absolute path to the image file.

`get_license_status`

read-onlyidempotent

Get license and trial status. Get the license and trial state before converting: registered or trial, clean outputs left (freeFilesRemaining), and what outputs become after the clean ones (afterFreeFiles: watermark). nextOutput says what the next output will be: clean or watermark.

No parameters.

`list_effects`

read-onlyidempotent

List effects. List all available image effects with their parameters.

No parameters.

`list_formats`

read-onlyidempotent

List supported formats. List the image formats Contenta Converter reads (extension, category, decoder) and writes (the CLI's contenta formats).

Parameter	Type	Required	Description
`kind`	string	no	Which list: input (formats read), output (formats written) or all (default). One of: `all`, `input`, `output`.

`merge_pdfs`

writes filescan replace files

Merge PDFs. Append multiple PDF files into a single PDF, in list order (page-level merge, no rasterization). During the trial the PDF always carries the trial watermark.

Parameter	Type	Required	Description
`input_paths`	array of string	yes	Required. Absolute paths to PDF files, appended in this order.
`output_path`	string	yes	Output PDF file path.

`read_metadata`

read-onlyidempotent

Read image metadata. Read EXIF, IPTC, and XMP metadata from an image file.

Parameter	Type	Required	Description
`input_path`	string	yes	Required. Absolute path to the image file.

`write_metadata`

writes filescan replace files

Write image metadata. Write EXIF/IPTC/XMP metadata to an existing image file without conversion. Every field goes to XMP too, so WebP, HEIC and AVIF (no IPTC block) keep keywords, city and country. The file is read back: written lists the fields now in it, dropped any that are not, with the reason.

Parameter	Type	Required	Description
`input_path`	string	yes	Required. Absolute path to the image file.
`copyright`	string	no	Copyright notice.
`creator`	string	no	Creator/photographer name.
`title`	string	no	Image title.
`description`	string	no	Image description/caption.
`keywords`	string	no	Keywords (comma-separated).
`rights`	string	no	Rights/usage terms.
`creator_city`	string	no	Creator city.
`creator_country`	string	no	Creator country.
`creator_email`	string	no	Creator email.
`creator_url`	string	no	Creator URL/website.
`latitude`	string	no	GPS latitude (decimal degrees).
`longitude`	string	no	GPS longitude (decimal degrees).
`altitude`	string	no	GPS altitude in meters.
`datetime`	string	no	Date/time override (ISO 8601).