Weigh the salt. Do not measure it. The salt ratio is the whole recipe: too little and spoilage organisms outcompete the lactobacillus you want, too much and the lactobacillus stalls and nothing sours. Kosher and fine sea salt differ enough in crystal size that a tablespoon of one is a different mass than a tablespoon of the other, which is why this card is written in grams and not spoons.
The source recipe disagreed with itself about the salt. It called for "3% to 3.5% total weight" but then gave 20-25 g for a 700 mL batch. Those are different numbers: 3.5% of the total (553 g of produce plus 700 g of water) is 44 g, while 20-25 g is 3-3.5% of the water alone. The stated grams and the stated method -- dissolve the salt in the water, pour the brine over the solids -- both describe a brine, so this card uses 24 g, a 3.5% brine. Both are safe; the 44 g version simply ferments slower and tastes saltier. Switch if that is what was meant.
Chlorinated tap water will not ferment. Chlorine and chloramine are in municipal water precisely because they kill bacteria, and they do not distinguish between the ones that spoil food and the lactobacillus doing the work here. Filtered, distilled, or boiled-and-cooled all solve it. Boiling drives off chlorine but not chloramine, so if your utility uses chloramine, filter it.
Everything solid stays under the brine. Lacto-fermentation is anaerobic -- the whole point of the airlock is to let CO2 out while keeping oxygen from getting in. A pepper floating above the liquid line is in air, and air is where mold and Kahm yeast grow. A glass weight costs a few dollars and is the difference between a jar you finish and a jar you throw out. Leave 1-2 in of headspace above that for the gas.
pH 4.6 is a safety threshold, not a preference. Below 4.6, Clostridium botulinum cannot grow, which is the reason acidity is what makes this shelf-safe rather than any cook step. A healthy pepper ferment lands at 3.5-3.8 on its own. Use a calibrated meter rather than paper strips -- strips are hard to read to a tenth, and a tenth is the resolution this decision needs. If it reads above 4.6, acidify with vinegar until it does not.
Swapping peppers does not change the arithmetic. Heat lives in the pepper choice, not in the brine, so jalapenos, serranos, habaneros, or a ghost pepper all drop into the same 450 g slot against the same 3.5% brine. Milder peppers or even sweet bells work the same way. Pick the Scoville level you want and leave the salt math alone.
The Kamado option changes what this is, in one direction only. Char the peppers on the grill BEFORE they go in the jar and you get smoke in the finished sauce with the ferment intact. Simmer or reduce the blended sauce outdoors AFTER and the heat kills the live cultures -- still good, no longer raw or probiotic, and it stops evolving in the bottle. That is a real trade, not a garnish, which is why the baseline here stays uncooked.
Why this is not filed as a Kamado cook. The grill step is optional on both ends, and the site filter is meant to say what a recipe IS. Same call as
Chicken Birria Tacos, where an optional grill sear does not earn a grill profile either. A jar on a pantry shelf is not a fire cook.