Short answer: you cannot certify a worn switch from a web page, but you can build the evidence a warranty or RMA reviewer actually wants — a repeatable, timestamped record of a single press producing two clicks, exported as a report you attach to the claim. A sentence like “my mouse double-clicks” is easy to wave away; a log showing the same fault across several runs, with the intervals in milliseconds and the browser and OS noted, is much harder to dismiss.
What counts as proof for a claim
Be precise about what you are proving. A browser cannot open the mouse or measure the switch, so it cannot certify the hardware. What it can do is record the symptom — real, timestamped click events — and let you show it is repeatable. For a claim, three things carry the weight: the doubled event on a click you made once, the short interval between the two clicks, and the fact that it happens again across separate runs rather than once by accident. That is symptom evidence, presented honestly, and it is usually enough to move a claim forward.
Build the evidence in the browser
- Open the double-click check and set a threshold — a value in milliseconds below which an adjacent click gets flagged. A few milliseconds is far shorter than any deliberate click, so keep the threshold low.
- Make at least five deliberate, separate single clicks on the affected button. Do not double-click on purpose — the whole point is that any second event came from the mouse, not from you.
- Read the run. It reports how many adjacent intervals fell below your threshold, and it is careful to call that a review filter, not a hardware fault count. A flagged interval is an event to look at, not a verdict.
- Run it several times, ideally across a couple of sessions. Chatter is often intermittent early on, so a fault that shows up on one run in three still counts — and repeatability is exactly what a reviewer looks for. The check keeps your runs in local History so you can build up a record.
If you want the plain-language version of the check first, the 60-second double-click check walks through a single run, and the chatter result reference explains what the numbers mean.
Export the report to attach to the claim
Once you have a few runs saved, open your History and use Copy report for a readable text summary, or Export JSON to download a cpskit-report.json file. Either one lists the runs, the threshold you used, the flagged intervals, and the browser and OS they were recorded on. Both include a built-in note that the results are browser-observed events, not a hardware certification.
That honesty is a feature, not a weakness. A report that overclaims (“this proves the switch is dead”) invites pushback; one that says exactly what it measured, and shows the fault repeating, reads as credible. Attach the exported file, and point to the open double-click input behavior dataset as a neutral reference for what the interval numbers mean.
What to write in the RMA or warranty message
Keep it factual and short. A reviewer skims, so lead with the fault:
- The symptom: a single press of the left button registers as two clicks.
- When it started and how often it happens now (“intermittent at first, now several times an hour”).
- The evidence: repeated runs of a browser double-click check show two clicks a few milliseconds apart from one press; report attached (
cpskit-report.json). - The environment the runs were recorded on — browser and OS — which the report already lists for you.
Do not guess at the cause in the message. “Two clicks from one press, repeatable, evidence attached” is stronger than “the switch is worn out,” which is a diagnosis you are not in a position to certify and which a reviewer may simply contest.
What this evidence can and cannot do
It confirms the doubled event and its timing, and it shows the fault is repeatable — that is the symptom, evidenced. It does not open the mouse, measure the microswitch, or certify a lifespan, and it says so on the report. It also does not measure polling rate, DPI, or click latency; a browser cannot see those, and a report that pretends to would hurt your claim, not help it. The general limits of browser input testing are laid out in how we test. The vendor may still run their own diagnostics, and this evidence supports a claim rather than guaranteeing it — but a repeatable, timestamped log is a far stronger starting point than a description in words.
The short intervals this check records are real button events observed in the browser; reading them as “chatter” is a symptom, not a hardware diagnosis, and the exported report states that limit itself. The open double-click input behavior dataset lists the reference numbers. On your machine the intervals will differ — what matters for a claim is a second event on a click you made only once, shown to repeat.