This is a sample of the kind of audit I do. It is based on an independent review I carried out on a public web game that is still in development. It was self-initiated and not done for a client, and I have left out names, brand art and anything that identifies the product. The findings are real. The game is a browser-based horse racing game where players connect a crypto wallet, race, earn the game's token and can stake it for rewards.
How I reviewed it
I read the full source code of the front end, checked the main screens against it, and compared what the interface says with what the code and the smart contract actually do. I did not assume that a feature works just because a screen for it exists. Where a finding comes only from reading the code and still needs a check in a browser, I say so.
The short version
The product has a strong starting point. The artwork has personality, the dark violet and glow feel premium, and the idea is easy to follow. The problems are about trust and consistency. Several screens promise things the product doesn't do, some failures are shown as normal states, and the website and the game feel like two different products. Most of this can be fixed with design decisions, before any new features are added.
The five things I would fix first:
- Visitors on tablets and phones cannot connect a wallet.
- The staking page promises a reward that the game doesn't apply.
- The results screen says rewards were sent before it knows they were.
- Failed loads look like empty or brand new accounts, and one pop-up can trap players.
- Buttons and sections on the landing page that lead nowhere.
What works
The character art, horses and farm setting give the game a clear personality. The colour palette is distinctive and easy to extend. The wallet button already handles several situations well, such as no wallet installed, wrong network and a rejected request. The first free races and the welcome bonus are a thoughtful way to onboard new players.
The findings
Each finding is marked Critical, High or Medium. Critical means it affects trust or blocks the main journey now.
The wallet can't be reached on tablets and phones
The wallet button is hidden below a certain screen width, but the mobile menu only appears at a much smaller width and doesn't include it. In between, and on phones, there is no way to connect. Every screen that needs a wallet becomes a dead end. This comes from the code and needs a browser check.
The staking screen, the contract and the game disagree
The staking page says the length of the lock sets the reward multiplier. The contract sets it by the amount staked, and the game reads the multiplier from an older system altogether. A player could lock tokens for a benefit they never receive. The team needs to choose one rule, and the design should show the player's real, current multiplier.
The results screen reports success too early
After a race it shows a green message that the reward was sent to the wallet, even if saving the race failed. The amount shown is an estimate worked out in the browser. Results need three honest states: pending, confirmed with a link to the transaction, and failed with a way to retry.
Failures look like zero
When the staking numbers can't be loaded, they show as zero. When the profile can't be loaded, it shows an anonymous new player with empty stats. Worst of all, the name-entry pop-up that new players see can't be closed, and it treats any server error as a new player. If the server is unavailable, the Save button stays disabled and nothing explains why. Every screen that loads data needs designed loading, empty and error states, with a retry.
A Buy tab that can't buy
The tab named Buy on the token page only shows information. There is no link to an exchange or swap. Players who arrive wanting the token have no way to get it.
Dead buttons and placeholder text
The first screen of the landing page has placeholder text and a Play Now button that does nothing. Several other buttons, the exchange logos, the social icons and the footer links also lead nowhere. On a token product every dead click costs trust. Each one should link somewhere or be marked as coming soon in the same way.
Choices that don't change anything
The pre-race screen asks players to pick a horse and a track, but only one of each is available. The selection isn't passed to the race at all. The entry fee is also taken before the controls are explained, and then there is a second Start button. Players should see how the race works first, pay once and start once.
A broken section after connecting
Once a wallet is connected, the token page shows an old NFT staking section with a spinner that never ends, zeros that never change, and a button that would fail. It should be removed until the feature exists.
One website built twice
The mobile version of the landing page is a separate copy of the page, and its wording has already drifted from the desktop one. One responsive design would be simpler, lighter and consistent.
No shared visual language
The code has around six different violets, three near-black backgrounds, six button styles and five typefaces, one of which appears in headings by accident. The game and profile screens use a different font and style from the marketing pages. A small set of colours, type styles and components would make everything feel designed together.
Missing basics
There is no page for errors or pages that don't exist. Many images have generic descriptions, keyboard focus isn't clearly designed, and the animations ignore reduced-motion settings.
What I would do first
A realistic first phase would be to agree the product rules that the screens depend on, such as the reward multiplier, how players get the token and which wallets are supported. Then fix the wallet access on every screen size, remove or label everything that doesn't work yet, design the loading, empty and error states, and put a simple design system in place. The visual redesign and new motion can follow once the foundations are in.
What an audit gives you
You get a written report like this one, with every finding tied to a screen and a priority. You also get a clear order of work and a realistic first phase you can hand to a designer or a developer.