With the multimeter set to VAC, you should see ~15-20 volts between all three leg combinations at 1000RPM idle. Revved up, it will build to 60+ VAC. All three combinations should be very nearly the same.
With the stator installed, set the multimeter to ohms, put one lead on ground, and then the other lead to each of the three legs. There should be NO circuit on any of them. If there's any continuity, you've got a problem (shorted stator).
Now, still on ohms, test the resistance between all three leg combinations. IIRC, the manual spec is between .4 and .8 ohms or so for each leg. My new one from Rick's Electrics measures about 1 ohm per leg and works perfectly fine though. If it's zero ohms, you've got a short somewhere, and if it's more than a couple ohms, you might have a burned pole.
However, stators are tough to test since they can short only under load, or only when hot, or both. My original checked out fine using the multimeter (good VAC, no short to ground, resistance right on), but I was only getting like 12.5 at the battery. I replaced the R/R with a new MOSFET style, and wired it straight into the stator, and the output straight to the battery (eliminating connectors). The voltage was good, until the motor got hot. Then back down to 12.5 again. Even testing the stator when hot, it still was OK. However I replaced the stator, now I have 13v hot idle, and 14.4 above 1500RPM or so.
Loose magnets don't sound particularly great either, though I have a hard time seeing that completely conking out your stator output. They'd have to be drastically out of place for their field to not effect the stator.