See your liquidation risk before the market does.
RiskLens scans the lending protocols you already borrow on, ranks the positions whose health factor is drifting into liquidatable territory, and tells you what that event would cost. It is read-only, non-custodial, and it never executes. No keys, no bot, no trade, just an early read of your own risk while you still have time to act on it.
Run a read-only risk scan See a sample risk read
THE PROBLEM
Leverage is fine until the day it is not, and on that day you usually find out from the liquidation engine, not from your own dashboard. A stablecoin wobbles, collateral drops, and a position you were not watching crosses its health-factor line while you are asleep. In October 2025 that pattern ran at scale (roughly nineteen billion dollars in leveraged positions liquidated in a cascade, with USDe depegging toward sixty-five cents), and the people who got hurt were not the ones who lacked a strategy, they were the ones who found out too late. The tools that watch these positions mostly exist to liquidate someone else faster. Almost none of them exist to let you see your own risk first.
HOW IT WORKS
RiskLens reads, and then it stops. You point it at your positions (an address, or a set of them), and it scans the protocols you actually borrow on, Aave, Compound, Morpho, Spark, for positions whose health factor has dropped toward the liquidation line. For each one it estimates the cost of the event (the profit at risk, net of the mechanics that make a liquidation expensive), and it ranks them so the thing most likely to hurt you is at the top. It never holds your funds, it never signs a transaction, and it never places a trade. Your keys stay yours, we only ever read the chain.
WHAT YOU GET (the risk read)
A ranked read you can act on, or hand to a risk committee: - every position drifting toward liquidation, sorted by severity, not buried in a feed, - the health factor and the distance to the line, so you can see how much room is left, - the cost of the event, the profit at risk quantified rather than left as a vague worry, - an auditable trail, a read you can show an auditor or a fund ops lead, not a screenshot of a Telegram alert.
A read you own, on positions you own, that no one executed against on your behalf.
WHO IT'S FOR
DeFi power users and small trading desks running leveraged positions who want an early-warning risk read they can trust, first, then crypto funds and treasuries that need a non-custodial, no-execution monitor they can put in front of a risk committee. If a liquidation you did not see coming can ruin your week, this is for you. If you want a bot to trade or auto-hedge for you, this is deliberately not that (and never will be under this brand).
PRICING (the ladder)
We price the ladder, not a single number, and every number below is a hypothesis we validate with you, not a commitment. - Single read, a one-off read-only scan of your positions, so you can see the ranked risk before you commit to anything (no money step, no execution). - Monitored, continuous watching with alerts as a position drifts toward the line, still read-only, still non-custodial. - Desk, portfolio-wide risk tiering across many positions with exportable, auditable reports for a risk committee.
IP is licensed, never assigned. No tier ever includes execution or custody. Any step that takes your money is gated and confirmed before it runs, nothing charges silently, and nothing ever touches your funds.
THE PROOF (dogfood)
We run the read-only discipline on ourselves before we sell it. The same scan path RiskLens uses to read your positions (a wired adapter seam, not a demo) is the one AYA uses across its own connect-and-use work, and every capability AYA ships terminates as a receipt on disk, the same shape of auditable read RiskLens hands you. And the product's own build left a receipt, the deliverable pattern was scored (92.75 out of 100, grade A, valid) and both of the atoms it composes were verified on disk before this page was written. The refusal to execute is not a missing feature, it is the design, and it is the same fail-closed posture (no custody, no outward action without a human in the loop) AYA applies to its own dangerous surfaces.
HONEST NOTE
We would rather under-promise. RiskLens has been built and structurally verified, the pattern composes, the atoms resolve, the scan path is a real wired handler. What it has NOT done yet is run against a live chain RPC or a real set of on-chain positions, that is the first live test once read-only chain access is wired, and we will say so plainly rather than imply a proof we have not run. The paired portfolio risk-tier span also has two honest bits of cleanup still owed (a placeholder reference to re-point and a duplicate copy to reconcile), both data edits, no new code, and we are not going to pretend otherwise. RiskLens reads, it ranks, it quantifies, and then it stops, and that stop is the whole point.
Run a read-only risk scan
*This page is a specification. The capability it describes is not built yet, and nothing here is a claim that it runs today.*