Skip to main content

Why is my scan slow?

First-time scans are slower because Qodex is still learning and authoring. During a first pass, Qodex may import endpoints, crawl pages, inspect flows, create scenarios, verify them, and save artifacts. After that work is saved, reruns should be much faster.

What to check first

Exploration is different from replay

Exploration creates tests. Qodex reads the surface, decides what matters, creates scenarios, and verifies them. Replay runs saved tests. API replay sends deterministic HTTP requests. UI replay uses saved browser actions and cached step resolution. If the first run is slow and the second run is fast, Qodex is behaving normally.

API runs

Saved API scenarios do not need LLM calls during normal replay. If an API replay is slow, the cause is usually the target API, network, database, or test environment. Check the endpoint timing before assuming Qodex is stuck.

UI runs

UI scenarios can be slower when the page changed or the replay cache is cold. Once Qodex has a stable action for a step, it can run that action directly in Chromium. If the UI changes and the cached action fails, Qodex may use recovery to resolve the original intent again.

When to investigate

Investigate when a saved scenario or stable suite stays slow after a successful first run. Do not worry if a first import, first crawl, or first scenario-authoring pass takes longer. That is where Qodex does the expensive thinking and setup.

Next steps

How Qodex works

See why discovery and replay behave differently.

Replay cache and self-healing

Learn how UI reruns get faster.

UI testing

Understand browser-driven testing.

LLM cost on reruns

See why reruns are cheaper than authoring.