You decided yesterday what deserves a seat in the suite; today you learn to read what the suite says back, because a run you cannot read is a run you cannot act on. Run everything you have and actually look at the output:
Running 12 tests using 4 workers
✓ verify that the practice store loads its catalogue (2.1s)
✓ verify that searching for jerk finds the Jerk Chicken Plate with its price (3.6s)
✓ verify that a 7 character password is rejected (2.4s)
✓ verify that the plantain price on the card matches the feed (3.9s)
- verify that removing a dish updates the total
✓ verify that adding a dish updates the count (3.1s)
1 skipped
11 passed (24.8s)
Read it like a tester. Eleven greens, and for each one the question you carried over from manual testing still applies: what did this actually prove. The dash is your quarantined cart test from Day 12: skipped, visible, carrying its ticket in the code. A skipped test that everyone can see is honest; a deleted one is amnesia.
The HTML report
Run npx playwright show-report. Every test is listed with every step inside it: each action, each locator, each wait, with timings. Click into your search test and find the 700 milliseconds the feed took; it is right there in the step timing. This is where you go first when something fails, and knowing the report while everything is green is what makes the red day calm instead of chaotic.
The trace: a filmstrip of the whole test
Now the deeper tool. Run with tracing on:
npx playwright test --trace on
npx playwright show-report
Open any test in the report and click its trace. You get a filmstrip of the entire run: every step with a screenshot before and after it, the network calls underneath, the console alongside. You can scrub back and forth through time and watch the store load, the search fire, the cards re-render. When a test fails in CI at 3 a.m., the trace is the difference between archaeology and a two-minute read. Learn it today, while nothing is wrong.
Run it twice
One more habit, and it is the cheapest health check a suite has: run the whole suite twice, back to back, changing nothing. If the greens repeat, good. A test that changes its answer between two identical runs is flake, and it is the subject of Day 17, but the habit of catching it starts now, because flake found by you is a fix, and flake found by the team is a credibility problem.
Read the report while it is green. The red day is not the day to learn where things are.
Do this today
Run the suite twice and compare. Open the HTML report and click through three tests, reading each step. Then run with —trace on and scrub through the filmstrip of your search test, watching the slow feed arrive. Tomorrow you break the suite on purpose, twice, and learn to diagnose reds fast.