Find the cause.
Prove the fix.
A deterministic tool suite and a fully structured AI audit pipeline — independent hunters, a jury, and a judge, running across multiple AI model families — go through your EVM-compatible contracts or your DAML source together, then every finding is correlated and deduplicated into one report your team can act on — not five separate opinions.
Unprotected mint()
No role or ownership check gates a supply-increasing function — any address can call it directly.
Every finding earns its place
Static analysis flags a lot. We only put a claim in front of you once it's been checked — by a second tool, a panel of reviewers, or a live replay.
Deterministic tool suite
Any EVM-compatible contract — Ethereum, Polygon, BNB Chain, Arbitrum, Optimism, Base. Slither, Aderyn, Mythril, Echidna, and Solhint read your Solidity together in an isolated sandbox and get reconciled into one report.
Structured, multi-model deep review
Five hunter lenses and a three-juror jury, running across independent AI model families, catch what static analysis alone misses — see the full pipeline below.
One list, not five
Tool output and AI-panel findings are deduplicated and correlated into a single findings list — no repeated claims, no manual reconciling.
Multi-round engagements
Fix, accept risk, or dispute each finding across rounds — the way a real audit firm re-checks work until the case closes.
Dig deeper or dispute
Push back on any AI-sourced finding — ask for a concrete PoC, or argue it's a false positive — and get a reasoned verdict.
Pay in USDC
One-click MetaMask checkout — no manual copy-paste, no card required.
A review built for DAML
Not an EVM audit with the names changed. Four risk axes and five lenses written for the language, every authorization finding replayed against a real Canton ledger, and the Daml compiler itself checking your upgrade against the version it replaces.
Your contract, through review, jury, and judge — before it's a report.
This isn't one model wearing five hats. Hunters, jurors, and the judge are routed across independent AI model families by design, so no single model's opinion ever becomes a finding on its own. Your contract passes through five independent reviewers, a three-juror panel, and a devil's advocate before anything is written into your report — deliberation only happens when the panel actually disagrees.
Source submitted
Your uploaded .sol or .daml files, or a fetched public GitHub repo, enter the pipeline as one combined source.
Five lenses, run concurrently
Each lens reads the same source looking for a different class of failure, and proposes candidate findings independently.
Three stances vote independently
Each juror votes cold. If all three agree, that's the verdict. If they split, each sees the others' anonymized reasoning and votes again.
A devil's advocate gets the final word
A separate reviewer actively tries to overturn any verdict the jury just confirmed — and can still demote it before it counts.
Correlated into your case file
One last completeness pass closes any gap, then tool and jury-confirmed findings are correlated, deduplicated, and compiled into your report.
A DAML audit is not an EVM audit with the names changed
Point an EVM audit at a DAML package and it finds nothing — there is no msg.sender, no gas, no reentrancy, no proxies for it to look at, and the failures that actually cost you here have no EVM equivalent at all. So we built the review DAML needs: four risk axes, five lenses, a jury, every authorization finding proven against a real Canton ledger, and — when you give us the version you are replacing — the Daml compiler itself checking your upgrade. Pick the DAML skill in the workspace and this is what runs.
We audit the failures DAML actually has
- The ledger computes authority and rejects an under-authorized submission before any code runs, so there is no
onlyOwnerto forget. The defect is authority wired to the wrong party, which the ledger then enforces faithfully, forever, without ever erroring - Privacy is the signature risk class — over-disclosure is only a defect on a shared ledger, and leaking to the wrong participant is frequently worse than an authorization slip
- Contention is a correctness property, not an operational one. One shared contract every participant must consume fails under ordinary honest load — the defect that works with one user and collapses with fifty
- And a fourth axis the EVM has no version of: what your next version would break for contracts already live on the ledger
Five lenses for a language the tools forgot
- There is no Slither for DAML. No Mythril, no Echidna, no Solhint — no static analyzer exists for the language at all, which is why most DAML packages go to production reviewed by nobody but the team that wrote them
- Five independent lenses read your package concurrently across three AI model families, then face the same three-juror panel and devil's advocate as an EVM audit
- Every authorization finding is replayed against a real, disposable Canton ledger — not a fork. A ledger that accepts a command it should have rejected is the proof
- Your package is compiled with
daml buildon Daml SDK 3.4 before any of it runs, so the review is against code that actually builds
Everything inside the workspace
Four areas, in the order you'll use them — the same four you'll see in the sidebar.
Submit the contract
Audit codeUpload .sol or .daml files, a whole folder, or fetch a public GitHub repo by URL. Pay in points or with one-click MetaMask.
Tools and AI review it
FindingsThe deterministic suite and the hunter/jury panel go through your source together, then everything is correlated into one list. Filter by severity, dig deeper on any finding, or dispute it.
Fix, justify, re-check
RemediationMark a finding fixed or accept the risk with a written reason. Submit updated source and every round is re-verified, so nothing closes unchecked.
Everything stays on record
HistoryAudits and the points they cost in one searchable timeline. Your source is never stored — we keep the record of the run, not your code.
Sign in with email, a connected wallet, or Google/GitHub — same account and history, whichever way you start.
What should we fix, and what should we run?
Both boxes go to a person. You do not need an account to use either — but if you are signed in, we can reply.
Tell us what is working, and what is not
If a finding was wrong, misleading, or missing, that is the most useful thing you can send us.
Request a model
The review panel runs several models against your contract. Tell us which one to add — we count the requests, and the models people actually ask for are the ones worth the integration work.
This one lives in the workspace. Requests are counted once per account, so we can tell how many people want a model rather than how many times a form was submitted — and asking again updates your note instead of adding a second vote.
Request a model in the workspace →