When we launched ⁠Discover, the first question from our own community was the right one: "how do we know the rankings aren't rigged?" Fair. This post is the answer — the exact rubric behind every verdict in the index, in enough detail that you could re-run the evaluation yourself and check our work. It's a condensed version of the full ⁠methodology page, with the reasoning behind each weight.

The five criteria

1. Update cadence (the heaviest weight)

Most scripts depend on game-specific timing — shot meters, recoil tables, animation windows. Live-service games patch those constantly, which means a script that isn't re-tuned is a script that's quietly decaying. We check when each script was last updated, how often it updates historically, and whether updates are included or cost extra. We wrote a whole post on ⁠why updates matter more than features; the rubric just turns that argument into a score.

2. Proof of gameplay

If a seller claims auto-green or aim assist and can't show uncut gameplay of it working, the claim is marked unverified. That standard applies to us too — it's why every yew.gg product page embeds live demos from ⁠/watch. Marketing renders and cropped clips don't count.

3. Support and setup

Zen setup has real failure points: firmware, connection order, PS5 authentication. The rubric checks what happens when a buyer gets stuck — is there a support channel with humans in it, or a forum thread from three years ago? Sellers with real Discord support and setup help score higher.

4. Pricing structure

Not the price — the structure. A $40 script with paid update DLC can cost more in a year than a $50 one-time script with free updates. Subscriptions aren't automatically bad, but the rubric asks one question: does the seller make money by keeping the script working, or by re-charging you when it breaks? Compare the ⁠lifetime and ⁠subscription categories and the difference is obvious fast.

5. Per-game feature depth

A "works with 100+ games" script usually means generic rapid fire and remaps everywhere, mastery nowhere. Game-specific depth — fade detection in 2K, per-weapon recoil tables in shooters — scores higher than breadth. This is why the ⁠NBA 2K26 category and the ⁠FPS category rank differently: the games reward different kinds of depth.

What the rubric deliberately ignores

• Popularity. Discord member counts are marketing, not quality.
• Feature-list length. Ten stale mods lose to four maintained ones.
• Who makes it. Our scripts get no home-field advantage — the ⁠Green breakdown lists its limitations in the same section every competitor's breakdown does.

Where the verdicts come from

Each listing's verdict is the rubric applied and then translated into plain English: who the script is actually for, and who should pick something else. When a free source is the right call — simple remaps, learning GPC — the verdict says "use the free one." That honesty is the entire value of the index. If we shade it once, the whole thing is worthless, and we know it.

Read the full Discover methodology

See the rankings — every script, every category