TaskSultan builds robotic process automation from code. You describe or demonstrate a process, the platform writes a TypeScript automation and the Robot runs it on a machine you control. The Studio, the diagram and the recorder are views of that code, never a second copy of it.
If you have not used the platform before, read What TaskSultan is and then Install. Both take a few minutes and leave you with something running.
#The parts of the platform
| Part | What it is for | Read |
|---|---|---|
| Studio | Write, run and watch automations in a browser IDE, or in the desktop shell | Studio |
| Recorder | Point at a screen element once and get a target the platform can find again | Recorder |
| Robot | Execute an automation, either from source or from a packaged build | Robot |
| Orchestrator | Queues, schedules, environments, credentials and the record of what ran | Orchestrator |
| Targets | The named elements your automations address | Targets |
| Framework | The transaction loop, the exception lanes and the run artifacts | Framework |
| Capabilities | Web, Excel, desktop, files, APIs, databases, email and PDF | Capabilities |
#Where to go next
New to the platform:
- What TaskSultan is gives you the shape of the thing in one page.
- Install gets the platform onto your machine.
- Your first automation builds and runs one.
Working on a specific job:
- Record a desktop process when the work happens in a fat client rather than a browser.
- Migrating from UiPath when you already have a Studio estate and want to know what converts.
- Credentials and the vault when a process needs to sign in somewhere.
#About these docs
Every page describes what the platform does today. Where something is unfinished, the page says so. Where two ways of doing a thing exist, the page says which one we recommend and why. The vocabulary is the product's own: a Target is a named element, a Transaction is one unit of work, a Run is one execution of one automation.