Build automation from scratch
A single automation handles one action—posting, scraping, remixing video. A deliverable ops plan usually needs a workflow that chains those pieces: collect related Facebook groups → remix video → generate product copy → publish automatically. This page walks that path with short videos—from one capability to a full plan—plus how to bind Android and iPhone devices.
Video tutorial
Watch 1 → 2 → 3 in order. The first two show how to build one automation from scratch; the third stitches scattered capabilities into a deliverable workflow. Videos 2 and 3 do not repeat every shared step from video 1—they focus on what is unique to each. Watch video 4 for Android, video 5 for iPhone.
1. Core flow: from recording to a running automation (Facebook group posting)
Start here. You will go through: record → pick an agent → create automation → develop in chat → adjust the create form → run for real. The next two videos build on this foundation.
2. Data-collection add-on (Facebook group scraping)
After posting, learn scraping. This video focuses on collection-specific settings and outputs; shared steps are lighter than in video 1—finish the core flow first.
3. Workflow: chain automations into an ops plan
Single automations are parts; a workflow is the deliverable vehicle. Demo path: collect related Facebook groups → remix video → generate product copy → auto-publish. After videos 1–2, each step’s role in the workflow will click.
4. Bind an Android physical device
The first three videos use Browser as the example. If you will run tasks on an Android phone, follow this video to bind the device and finish environment setup. After that, you can cast and run tasks.
5. Bind an iPhone
The first three videos use Browser as the example. If you will run tasks on an iPhone, follow this video to bind the device and finish environment setup. After that, you can cast and run tasks.
Before you start
Still using Browser as the example:
- Finish Quick start (client installed and signed in; download first if needed)
- Have a fingerprint browser you can start (Platform browser is fine)
- Before recording, turn off top-right Tasks (task execution master switch; some recording copy says “task running”)
Single automation: flow overview
Matches videos 1 / 2—turn one recording into a runnable automation.
| Step | What you do in the client | Outcome |
|---|---|---|
| 1 | Recording → create → add track → finish | Completed reference recording |
| 2 | Agents → Browser automation expert → Create automation wizard | Pick recording + project ID → start chat |
| 3 | Chat: “Develop from recording and trial-run script” | Script runs |
| 4 | Chat: adjust create form + Create preview | Task-group form ready |
| 5 | My automations → card → confirm submit → enable Tasks | Production batch run |
Key notes
- A recording captures how a human operates—it is not the final script. Only Completed recordings that match the device type and are not yet bound to another automation can be selected.
- When building from scratch, picking an RPA expert template goes straight into the Create automation wizard; the workspace is handled automatically.
- Development chat uses Development mode. For a failed task group, use Task group management → Investigate and fix (Troubleshooting mode). The expert button on the card continues script changes.
- Create preview in chat is convenient (refreshes ~3s after config changes). For production batch runs, create task groups from My automations by clicking the card body (not the expert button).
- Production task-group runs usually use almost no tokens; chatting for changes or fixes is what consumes tokens.
- To stitch collection, creation, and publishing into one deliverable plan, use a workflow (see video 3 and Workflows) instead of cramming everything into one automation.
Related
- Quick start
- Devices & automation
- Workflows: multi-step ops plans
- Case Studies · Download · Register · Sign in