Web3
Wallet Scrutiny
Open-source security platform verifying Bitcoin wallet transparency for a safer Web3 ecosystem.
- Role
- Product Designer
- Timeline
- 2025 · 2 months
- Industry
- Web3
- Platform
- Web, desktop and mobile
A very specific kind of trust
Wallet Scrutiny is an open-source project that reviews Bitcoin wallets for one very specific risk: could the people behind a wallet take every user’s funds at once, without anyone noticing in time? To answer that, it runs a sequence of tests (is it really a wallet, can you hold your own keys, is the source code public, can the published app be reproduced from that code), and it now relies on community members reproducing builds and publishing signed results.
It’s careful, rigorous work. It was also, for most visitors, hard to read. The people who most need the answer, everyday Bitcoin users choosing a wallet, are the least likely to know what “reproducible build” means.
Not everyone is a developer
The evidence was already there. The problem was translation: turning a technical verdict into something a non-developer could understand, compare and act on, without dumbing it down or overstating what the review proves.
That second half matters. Wallet Scrutiny is explicit that it isn’t a security audit and doesn’t endorse any wallet. So the design couldn’t turn verdicts into green badges that read as “safe”. It had to make results understandable while keeping their meaning exactly as narrow as it is.
What the redesign had to do
- Get someone from “I use this wallet” to its verdict in one step.
- Make the result readable at a glance, and the reasoning one click deeper.
- Keep the language honest: progress and evidence, not endorsement.
- Work as well on a phone as on a desktop, because that’s where most wallets live.
From verdicts to evidence
- The promise, in human terms“Know your wallet like you built it.” The line under it, “Not everyone is a developer. Not everyone has to be.”, names exactly who this is for.
- Search is the front doorMost visitors arrive with one wallet in mind, so the page leads with search across thousands of reviews, not an explanation of the methodology.
- Filters speak in outcomes“All 10 tests passed” and “All operating systems” instead of internal category names.
- A ten-dot meter“Passed tests 1–5 of 10” shows how far a wallet got through the sequence: precise, comparable, and never claiming more than the review does.
- Risks that aren’t pass/failLabels like “Vaporware” and “Not updated in a long time” surface what doesn’t fit on the meter.
- Same hierarchy on mobilePromise, search, then filters as full-width controls in a single column.
Progress, not a score. A single rating would have been easier to scan and wrong in spirit. The tests run in order, and a wallet that fails early never reaches the later ones, so the meter shows how far a wallet got. It’s comparable at a glance and true to the methodology.
Plain words on the surface, exact words underneath. Cards say “Passed all 10 tests”; the wallet page explains custody, source availability and reproducibility in full. Nobody has to learn the vocabulary to get the answer, and nobody who already knows it loses precision.
Each card gets its own colour, softly. Wallets are recognised by their icons, so each card picks up a pale tint that echoes the app’s brand. It makes a grid of thousands feel browsable, while the verdict stays in the same neutral place on every card.
Desktop and phone
Still on the homepage
The framing outlived the project. The live site still opens with “Know your wallet like you built it. Not everyone is a developer. Not everyone has to be.”, still leads with search, and still describes results as “Passed all 10 tests”. The product has kept growing (community verifications, signed results), but the way it speaks to non-technical users is the one this work set up.
Honesty is a design constraint
The temptation with a trust product is to make it reassuring. This one taught me that the more respectful move is to make it legible: show exactly what was checked, how far a wallet got, and what that does and doesn’t mean. Then let people decide.
Translating technical evidence for non-experts isn’t simplification. It’s editing, very carefully.
Got a user problem?
Open to interesting product problems, collaborations and AI-driven ideas.
or write to obianujuohuegbe@gmail.com