62/ 100

At risk

Suite health

78.2% of executions passed across the analysed window · 19 Jan 2026 – 7 Feb 2026

Runs analysed: 20
Tests tracked: 231
Executions: 4,620
Browsers: 3

Latest (Run 20): 231 tests — 171 passed, 54 failed, 6 skipped (74% pass rate).

Suite summary

Every one of the 231 tracked tests, by how it behaves across 20 runs.

Total tests

231

Passing

144

Passing on retry

Flaky

Newly failing

Consistently failing

Skipped

Passing consistently14462.3%
Flaky2812.1%
Consistently failing3414.7%
Newly failing166.9%
Recently fixed31.3%
Skipped62.6%

Flaky tests over time

Tests whose outcome alternates, counted per run.

28now · ranged 0–28 across the window

Show values

Run	flaky tests
Run 1	0
Run 2	0
Run 3	25
Run 4	25
Run 5	25
Run 6	25
Run 7	25
Run 8	25
Run 9	25
Run 10	25
Run 11	25
Run 12	25
Run 13	25
Run 14	25
Run 15	25
Run 16	25
Run 17	25
Run 18	25
Run 19	28
Run 20	28

Up 28 since the first run — instability is spreading to more tests.

Retries per run

How hard the suite worked to recover, run by run.

47in the latest run · 41 average · 819 total

Show values

Run	retries
Run 1	25
Run 2	50
Run 3	25
Run 4	47
Run 5	46
Run 6	47
Run 7	46
Run 8	47
Run 9	25
Run 10	50
Run 11	25
Run 12	47
Run 13	46
Run 14	47
Run 15	46
Run 16	47
Run 17	25
Run 18	56
Run 19	25
Run 20	47

Swings between 25 and 56 per run — retry load is uneven, so some runs are recovering far more than others.

Pass rate over the run window

Every analysed run, oldest first. Hover for that run's figures.

Show run-by-run figures

Run	Pass rate	Failed	Flaky	Retries
Run 1	81.4%	40	0	25
Run 2	71.9%	62	0	50
Run 3	82.7%	37	25	25
Run 4	81%	41	25	47
Run 5	73.6%	58	25	46
Run 6	81%	41	25	47
Run 7	73.6%	58	25	46
Run 8	81%	41	25	47
Run 9	82.7%	37	25	25
Run 10	71.9%	62	25	50
Run 11	82.7%	37	25	25
Run 12	81%	41	25	47
Run 13	73.6%	58	25	46
Run 14	81%	41	25	47
Run 15	73.6%	58	25	46
Run 16	81%	41	25	47
Run 17	82.7%	37	25	25
Run 18	70.6%	65	25	56
Run 19	82.7%	37	28	25
Run 20	74%	54	28	47

What the analyzer recommends

Generated by the deterministic engine, in its own words.

Pass rate is below 80%. Review recent changes.
High
34 stable failure(s) consistently failing. Prioritize these.
High
28 flaky test(s) found. Review retry configurations and test isolation.
Medium

Run highlights

The facts behind the numbers above.

Suite scope
231 tests · 20 runs
144 passed in every run.
Latest run (Run 20)
74% pass rate
171 passed, 54 failed, 6 skipped.
Flaky tests
28 detected
Outcomes alternate between pass and fail across the window.
Newly failing
16 tests
Passed earlier in the window and are failing now.
Consistently failing
34 tests
Failed in every run — reproducible, not intermittent.
Dominant failure category
Network
426 of the classified failures (24.1%) — the largest single concentration.
Retry behaviour
819 retries
3 test(s) recovered on retry; the rest failed again.
Browser spread
webkit worst at 24.4%
chromium is lowest at 18.5% — a 5.9 point spread.
Awaiting human review
18 failures
The engine declined to name a cause and routed these to a person.

Failure intelligence

20 deterministic rules analyze failure patterns before AI investigation.

Signal	Category	Tests	Confidence	Next step
Needs manual investigation Could not confidently determine a specific root cause — this failure needs a manual look chromium, firefox, webkit	Unknown	9	27%	Needs human review
Test is consistently broken — not flaky, genuinely failing chromium, firefox, webkit	Stability	6	99%	Needs human review
Network request failed at the protocol level — unreachable host, DNS failure, or TLS error chromium, firefox, webkit	Network	6	98%	Evidence supports this signal
Authentication required — the request was not authenticated chromium, firefox, webkit	Authentication	6	86%	Evidence supports this signal
A network request failed — API unreachable, timeout, or DNS failure chromium, firefox, webkit	Network	6	78%	Evidence supports this signal
The locator did not match any element in the DOM chromium, firefox, webkit	Locator	4	91%	Evidence supports this signal
Page or API response too slow chromium, firefox, webkit	Timeout	4	88%	Evidence supports this signal
Element exists in the DOM but is not visible — timing or rendering issue chromium, firefox, webkit	Locator	4	78%	Evidence supports this signal
The locator matched multiple elements — Playwright's strict mode requires a single match chromium, firefox, webkit	Locator	3	99%	Evidence supports this signal
Authorization denied — the authenticated user lacks required permissions chromium, firefox, webkit	Authentication	3	94%	Evidence supports this signal
The requested endpoint or page was not found chromium, firefox, webkit	Network	3	94%	Evidence supports this signal
Internal server error — the backend encountered an unexpected condition chromium, firefox, webkit	Network	3	94%	Evidence supports this signal
Expected title does not match actual page title chromium, firefox, webkit	Assertion	3	92%	Evidence supports this signal
The browser or page was unexpectedly closed during test execution chromium, firefox, webkit	Environment	3	89%	Evidence supports this signal
Expected value did not match actual value — test assertion failed chromium, firefox, webkit	Assertion	3	87%	Evidence supports this signal
Element was present but not ready for interaction when clicked chromium, firefox, webkit	Locator	3	82%	Evidence supports this signal
The network connection was abruptly terminated by the server or network chromium, firefox, webkit	Network	3	77%	Evidence supports this signal
Race condition or async timing issue — the test outcome depends on execution order chromium, firefox, webkit	Stability	3	67%	Needs human review
The element was removed from the DOM and re-rendered between location and interaction webkit	Locator	2	81%	Evidence supports this signal
Expected text does not match actual text on the page firefox	Assertion	1	85%	Evidence supports this signal

3 of 20 rules need human review because the available evidence isn't strong enough to support a conclusion.

View all 20 rules →Investigate 78 failures →

Featured investigations

Three failures showing three different investigation outcomes.

Flaky

Checkout — declined card retry

A payment confirmation test that fails, then passes when retried.

7 of 20 tracked runs failed · 14 flipsInvestigate →

Consistently failing

Billing — monthly invoice generation

An invoice test that fails on every run with a server error.

20 of 20 tracked runs failedInvestigate →

Newly failing

Settings — theme preference

A test that failed recently, with very little evidence captured.

1 of 2 tracked runs failed · 1 flipInvestigate →

Deterministic analysis

Runs before the AI, on every failure.

Rules fired

20/20

Failures analysed

Distinct failure signatures28
Escalated to human review18
Most common categoryNetwork

See all 20 detection rules →

Failure categories

How failures classify.

Element / selector2221.8%
Network2120.8%
Unclassified1615.8%
Backend / API1514.9%
Authentication98.9%
Assertion76.9%

Failure rate by browser

Share of executions that failed.

webkit27.3%21 of 77
firefox22.1%17 of 77
chromium20.8%16 of 77

Retries recorded819

Recovered on retry3 tests