Skip to content

README.md

Latest commit

 

History

History
344 lines (240 loc) · 9.67 KB

README.md

File metadata and controls

344 lines (240 loc) · 9.67 KB
Raw

Languages: English | 简体中文 | 繁體中文 | Français | Español | Русский | 日本語

deckops CLI

Deckops is the command-line tool for Deckflow. Use it to upload files, create async tasks, and check task status.

Requirements

  • Node.js >= 18
  • A valid Deckflow account and workspace (space)

Installation

npm install -g deckops
deckops --help

Quick start

Deckops will guide you through authentication and workspace setup when a command needs it.

deckops convert slides.pptx --to pdf

Global options

Option Description
--json Output results as JSON (ideal for scripting)
--version Show version number
--help Show help information

Examples:

# List tasks in JSON mode
deckops --json task list --limit 5

# View version
deckops --version

# View help for a subcommand
deckops convert --help

File compression

compress <input-file>

Compress Office documents, videos, or zip files. Task type is selected automatically based on file extension.

Supported formats: .zip, .pptx, .key, .docx, .xlsx, .mp4, .avi, .mov, .mkv

Option Description
-o, --out <path> Write task result to a file or directory
--no-wait Do not wait for completion after creating the task
--timeout <seconds> Wait timeout (default 300 seconds) 
# Compress a PPT presentation
deckops compress presentation.pptx

# Compress video and save result
deckops compress demo.mp4 -o ./output/compressed.mp4

# Create task only, do not wait
deckops compress large.pptx --no-wait

Information extraction

extract <input-file>

Extract fonts, text shapes, and other information from a file.

Option Description
--type <type> Extraction type: fonts, text-shapes
-o, --out <path> Save result
--no-wait Do not wait for completion
--timeout <seconds> Timeout in seconds 
# Auto-extract font info from pptx
deckops extract slides.pptx

# Explicitly extract text shapes
deckops extract slides.pptx --type text-shapes

# Extract and save to directory
deckops extract slides.pptx --type fonts -o ./extracted/

OCR text recognition

ocr <input-file>

Perform OCR on images. Supports .jpg, .jpeg, .png.

Option Description
--language <lang> Language (default zh-hans)
-o, --out <path> Save result
--no-wait Do not wait for completion
--timeout <seconds> Timeout in seconds

Supported languages: zh-hans, zh-hant, en, ja, ko, ar, de, es, fr, it, pt, ru

# Recognize Chinese image (default language)
deckops ocr scan.jpg

# Recognize English image
deckops ocr document.png --language en

# Recognize Japanese and save
deckops ocr receipt.jpg --language ja -o ./ocr-result.json

Format conversion

convert <input-files...> --to <format>

Convert files to the specified format.

Option Description
--to <format> Output format (required): image, pdf, video, html, png, pptx, webp
--width <number> Width for HTML → PPT/PNG
--height <number> Height for HTML → PPT/PNG
--need-embed-fonts [boolean] Whether to embed fonts for HTML → PPTX (default false)
-o, --out <path> Save result
--no-wait Do not wait for completion
--timeout <seconds> Timeout in seconds

Multi-file note: Only html → pptx supports multiple input files, merged in order into a single conversion task.

# PPT to PDF
deckops convert slides.pptx --to pdf

# Merge multiple HTML files into PPTX
deckops convert page1.html page2.html page3.html --to pptx

# HTML to PNG with dimensions and save
deckops convert slide.html --to png --width 1920 --height 1080 -o ./slide.png

PPT merge

join <input-files...>

Merge multiple .pptx files in the given order into one (task type pptx.join). At least 2 files required.

Option Description
--name <name> Task name
-o, --out <path> Save result
--no-wait Do not wait for completion
--timeout <seconds> Timeout in seconds 
# Merge three presentations
deckops join intro.pptx body.pptx appendix.pptx

# Specify output task name and save
deckops join part1.pptx part2.pptx --name merged-deck -o ./merged.pptx

# Create task only
deckops join a.pptx b.pptx --no-wait

AI content generation

create [input-files...]

Generate document content from text or reference files (API task type generation).

Option Description
--input-text <text> Input text
--enable-search [boolean] Enable search
--advanced-model [boolean] Use advanced model
--fast-mode [boolean] Fast mode
--intent <intent> Generation intent
--audience <audience> Target audience
--page-count <number> Expected page count
--author <name> Document author
-o, --out <path> Save result
--no-wait Do not wait for completion
--timeout <seconds> Timeout in seconds

Up to 2 reference files supported: .html, .pdf, .docx, .pptx, .txt, .md, .mm, .xmind, .ipynb

# Pure text generation
deckops create --input-text "Write a product launch plan"

# With reference file and page limit
deckops create outline.md --input-text "Expand into a full speech" --page-count 20

# Advanced model + search, and save result
deckops create brief.pdf --input-text "Generate a detailed report" --advanced-model --enable-search -o ./report/

Document translation

translate <input-file>

Translate document files. Supports .docx, .pptx, .pdf, .xlsx, .key.

Option Description
--from <language> Source language (required)
--to <language> Target language (required)
--model <model> Model (required): Standard or Pro
--use-glossary [boolean] Use glossary
--image-translate [boolean] Translate text in images
-o, --out <path> Save result
--no-wait Do not wait for completion
--timeout <seconds> Timeout in seconds 
# Chinese to English (Standard model)
deckops translate handbook.docx --from zh --to en --model Standard

# English to Chinese, Pro model
deckops translate slides.pptx --from en --to zh --model Pro

# Auto-detect source language, enable glossary and save
deckops translate manual.pdf --from auto --to ja --model Pro --use-glossary -o ./translated.pdf

Generic task execution

run <task-type> <input-files...>

Execute with an explicit task type. Suitable for advanced usage or tasks not wrapped by the CLI.

Option Description
--param <key=value> Task parameters (repeatable; values support JSON)
-o, --out <path> Save result
--no-wait Do not wait for completion
--timeout <seconds> Timeout in seconds

Multi-file note: The following task types support multiple input files as an ordered source set: pptx.join, convertor.html2pptx, html.buildPlayer, generation.

# PPT to PDF (explicit task type)
deckops run convertor.ppt2pdf demo.ppt

# Merge PPT
deckops run pptx.join part1.pptx part2.pptx

# HTML to PPTX with parameters
deckops run convertor.html2pptx page1.html page2.html --param width=1920 --param needEmbedFonts=true

Task management

task list

List tasks in the workspace.

Option Description
--type <type> Filter by task type
--limit <n> Maximum count (default 50)
--offset <n> Pagination offset (default 0) 
# List 10 most recent tasks
deckops task list --limit 10

# Conversion tasks only
deckops task list --type convertor.ppt2pdf --limit 20

# Paginated JSON query
deckops --json task list --offset 50 --limit 50

task get <task-id>

Get details for a single task.

Option Description
-o, --out <path> Download and save result of completed task 
# View task details
deckops task get abc123-task-id

# Download task result to file
deckops task get abc123-task-id -o ./result.pdf

# JSON mode
deckops --json task get abc123-task-id

task delete <task-id>

Delete a specified task.

# Delete task
deckops task delete abc123-task-id

# JSON mode
deckops --json task delete abc123-task-id

# Confirm list after deletion
deckops task delete abc123-task-id && deckops task list --limit 5

Interactive completion

In a TTY environment, if required arguments or options are missing, the CLI will attempt interactive prompts (not triggered when using --json). For example, when --to is missing, convert will guide you to choose an output format.

Output and -o behavior

  • Default: human-readable text + progress spinner
  • --json: structured JSON for script parsing
  • -o, --out: automatically download result after task completion
    • Single file → write to specified path
    • Multiple files → write to directory; if path ends with .zip, pack as zip
    • No file result → write JSON

When -o is specified, the CLI waits for task completion so it can download the result, even if --no-wait is also provided.

FAQ

Q: When is multi-file input valid?

Only some tasks support ordered multi-source: convert for html → pptx, join, create (up to 2 reference files), and run for pptx.join / convertor.html2pptx, etc.

Related links