Skip to content

QA-PW-141 — Retries configured without a flake-triage loop ​

← Back to the rule catalog

Generated from the rule registry. Get the same detail from the CLI with mjolnir explain QA-PW-141.

FieldValue
Severitywarning
Confidencehigh
Tierextended
Measured FP rate9% (n=33)
Evidence levelE1
QA impactFalse-green risk (FALSE-GREEN)
False-positive risk (author estimate)low
Autofix availableno
Languagestypescript, javascript
Frameworksplaywright
Detection strategyLEXICAL (regex heuristic)
Introduced inv0.3.8

Why this fails in production ​

Retries convert intermittent failures into silent passes. Without a forensics/triage step consuming retry data, flaky tests pass forever and real regressions hide behind lucky reruns.

What gets flagged (real detector output) ​

retries: undefined with no visible flake-triage loop.

Example from this rule's own must-fire fixture: tests/fixtures/QA-PW-141/must-fire/playwright.config.ts

The fix ​

Keep retries <= 2 and feed retry outcomes into the evidence path (explain --evidence), or add a reporter so flaky passes are reviewed.

Confirmed NOT to fire on the corresponding clean pattern ​

Verified against tests/fixtures/QA-PW-141/must-not-fire/playwright.config.ts — a legitimate, similar-looking pattern this rule correctly leaves alone.

Corpus-measured false-positive risk ​

Real occurrence counts from npm run corpus:audit against actively-maintained OSS repos — reproduce yourself, don't just trust this table (see the false-positive audit):

RepoOccurrences
calcom-cal2
dubinc-dub1
grafana-grafana1
hashicorp-vault1
negative-fixtures24
nextauthjs-next-auth1
positive-fixtures24
vercel-next-js1
vitest-dev-vitest1
withastro-astro2

Released under the MIT License.