When Luotsi Fits
Use this page when the main question is not how Luotsi works, but whether it is the right kind of tool for the job.
Luotsi is usually a good fit when
Section titled “Luotsi is usually a good fit when”- The workflow is built around real Android devices over adb rather than a browser DOM.
- You want live view, structured inspection, repeatable scenario runs, and replay artifacts in one stack.
- Engineers, agents, and CI need to share the same command surface and output contracts.
- Post-run evidence matters. Screenshots, hierarchy dumps, logcat, telemetry, and replay summaries need to survive the session.
- The same workflow should make sense on a laptop, a shared lab host, and a CI runner.
Luotsi is usually a weak fit when
Section titled “Luotsi is usually a weak fit when”- The real problem is browser automation, not Android device automation.
- You want a managed cloud platform to own both the devices and the orchestration model.
- The team expects recorder-first authoring instead of explicit JSON playbooks and command-driven flows.
- You need immediate cross-platform coverage, especially iOS, from the same tool.
Quick self-check
Section titled “Quick self-check”If most of these are true, Luotsi likely fits:
- You already have or can get adb access to the devices that matter.
- You want machine-readable JSON or JSONL rather than a UI-only control path.
- You want failure triage to continue after the live device session ends.
- You want one workflow that can move from local exploration into CI or a shared lab.
Choose the next page by the job
Section titled “Choose the next page by the job”- Use Android Automation For Engineering Leads if the decision is about team architecture, rollout, or evaluation scope.
- Use Luotsi Alternatives And Comparison if the decision is really about tradeoffs versus browser automation, device farms, or recorder-style tools.
- Use AI Agent Android Automation if the main question is whether an agent loop can drive the workflow.
- Use Android CI Device Lab Workflows if the pressure comes from shared infrastructure and pipeline reliability.
- Use Replay-Driven Triage if the real bottleneck is what happens after a failed run.
Recommended first commands
Section titled “Recommended first commands”luotsi doctor --device <serial>luotsi inspect --device <serial>luotsi scenario-validate --path scenariosluotsi run --path scenarios --device <serial>Those commands cover the shortest path from device readiness to exploration and then into repeatable automation.