The Node SDK now ships with Intent commands: high-level CLI commands for common Transloadit tasks. They make it much easier to reach for Transloadit for one-off jobs, without first writing Assembly Instructions JSON by hand.

Social preview image for the Node SDK Intents announcement.

This grew out of a practical problem on our side. When we are building lasting integrations, JSON instructions and reusable templates are exactly the right abstraction. They are expressive, stable, and well suited to custom multi-step workflows. For quick command-line jobs, though, they can feel like overhead.

Intents close that gap with guessable commands such as markdown pdf, markdown docx, image remove-background, image generate, and image describe.

Run one-off jobs from the CLI

If your Transloadit credentials are available in the shell:

export TRANSLOADIT_KEY=MY_AUTH_KEY
export TRANSLOADIT_SECRET=MY_SECRET_KEY

you can now run:

npx -y transloadit markdown pdf --input ~/code/uppy/README.md --print-urls

And get output like:

STEP     NAME        URL
convert  README.pdf  https://pub-9974528f53db4a839e95f7e6ce9adb68.r2.dev/836b8ecaed114779b02cb99d4540d81f/37053087fe38475ea3fba78d87dbdfd3/9af5485747b446c5ab737b64db23afc4.pdf

That means you can use Transloadit as a practical CLI tool for quick jobs, not only as an embedded back-end integration.

Use high-level commands instead of JSON for common tasks

Intents cover common workflows with short, predictable command names. A few examples:

npx -y transloadit image remove-background --input ./photo.jpg --print-urls
npx -y transloadit image generate --prompt "A red bicycle in a studio" --out bicycle.png
npx -y transloadit image describe --input ./hero.jpg --for wordpress --out fields.json
npx -y transloadit markdown docx --input ./plan.md --out plan.docx

The idea is simple: for the most common use cases, the fastest path should be a command you can guess.

That also now includes input-guided image generation. For example:

npx -y transloadit image generate \
  --input ./person1.jpg \
  --input ./person2.jpg \
  --input ./background.jpg \
  --prompt "Place person1.jpg feeding person2.jpg in front of background.jpg" \
  --out scene.png

By default, image generate now uses google/nano-banana-2, so prompt-only and input-guided runs both start from the same stronger default model.

This is especially useful when:

  • you are solving a quick one-off processing task
  • you are working over SSH, in tmux, or in CI
  • you want a result URL immediately via --print-urls
  • you are driving the workflow from a coding agent or LLM

For the full reference, see the Intent Commands reference on GitHub or the Node SDK docs.

One useful detail is --print-urls. It is designed for environments where you may not want to manage local output paths at all.

Instead of forcing a local --out, the CLI can print temporary result URLs directly:

npx -y transloadit markdown pdf --input ./README.md --print-urls

This works well in remote shells, inside automation, and in agentic workflows where a URL is the fastest hand-off to the next step.

Use Intents for quick jobs, and JSON for custom workflows

Intents are not a replacement for Transloadit's lower-level model. They are the fast path for common tasks.

Use Intents when you want:

  • a quick conversion or transformation
  • a command that is easy to discover from the terminal
  • a better fit for scripting, automation, or LLMs

Use full JSON instructions or reusable templates when you want:

  • custom multi-step pipelines
  • fine-grained Step configuration
  • long-lived embedded workflows inside your product

That distinction matters. We still think JSON instructions are the right long-term abstraction for custom media pipelines. Intents simply make the short path shorter.

Make Transloadit easier to reach for, including from LLMs

Another motivation here was to make Transloadit easier to use from newer agent-driven workflows. When a developer or an LLM can guess markdown pdf or image describe, they are more likely to actually reach for Transloadit instead of stitching together ad hoc local tools.

That is good for developer experience, but it also helps us expose the same platform capabilities in a more natural way across CLI usage, automation, and AI-agent environments.

We also updated our public skills repo so agents can use the same Intent-based flows directly, including the new multi-input image generate example.

Try it

If you already have a Transloadit account, export your credentials and run:

npx -y transloadit --help

Then try one of the new commands on a real file.

For more details, see:

If you outgrow the high-level commands, you can always drop down to full Assembly Instructions and templates later. That path still exists. It is just no longer the only sensible starting point from the CLI.