Do applications always expose Jev directly to users?

Usually no. Users see outcomes (ranked results, blocked actions). Jev runs server-side with thresholds set by the product team.

What is the every CLI?

every scans functions in a codebase and scores how well each matches a natural-language yes/no question using Jev, caching hashes locally. See the every listing for self-test metrics and API key requirements.

Can I submit my shipped app?

Yes. Use Submit on this site with a repo or demo link and a short note on how Jev primitives map to your UX.

How do apps differ from playgrounds?

Playgrounds teach primitives interactively. Applications target a recurring job (moderate, triage, trade) even if they also ship a demo UI.

madbid.lol · The paid leaderboard for AI builders. Products, agents and agencies buy their rank.

Applications

Shipped products: trading, moderation, analytics, triage, RAG verify, support pipelines.

Things people run in production (or at least on mainnet demos). Moderation, trading, triage, rerankers. Application listings skew toward user-visible outcomes: blocked comments, routed tickets, approved refunds, verified citations. They are the best proof that System One patterns survive outside research repos. Check each README for deployment and data handling before you send real traffic.

Who it is for: Teams looking for proof that typed decisions survive outside agent accessories: customer-facing moderation, finance, support, or developer tools with real traffic stories.

How to choose: Read deployment notes and data handling first. Favor apps that show how thresholds map to user-visible outcomes (blocked comment, ranked function, approved trade). Demos with live URLs beat screenshot-only repos.

75 listings in this bucket · Learn Jev

Guide

Applications vs agent tooling

Agent tooling sits beside your coding agent. Applications are products where end users or downstream systems feel the Jev decision: a ranked inbox, a blocked URL, a moderated comment, a ranked function list. They are the best evidence that System One patterns work when UX depends on them.

Highlights

Hand-picked listings with extra detail on this directory. Star counts and demos still live on each card.

every
Ask a yes/no question of every function in a repo; Jev scores each unit (function-level search).
jev-trader
Trading automation with discrete Jev choices.
notra
Note capture with structured Jev triage.
jevmeter
Instrument latency and calibration of your own Jev calls.

FAQ

Do applications always expose Jev directly to users?: Usually no. Users see outcomes (ranked results, blocked actions). Jev runs server-side with thresholds set by the product team.
What is the every CLI?: every scans functions in a codebase and scores how well each matches a natural-language yes/no question using Jev, caching hashes locally. See the every listing for self-test metrics and API key requirements.
Can I submit my shipped app?: Yes. Use Submit on this site with a repo or demo link and a short note on how Jev primitives map to your UX.
How do apps differ from playgrounds?: Playgrounds teach primitives interactively. Applications target a recurring job (moderate, triage, trade) even if they also ship a demo UI.

All listings

ApplicationsTypeScript794

DemoFeatured

jev-trader

One AI trade call every Monad block on Kuru MON-USDC, with Jev picking the move instead of a market essay.

by @jarrodwatts

applicationstrading