A DeFi lending protocol has different failure modes than an NFT marketplace. A wallet contract can't be tested like a GameFi rewards system. We don't run the same checklist on every contract and call it done—we shift our threat model to match what actually breaks in your project type.
DeFi protocols: Most DeFi exploits are economic, not technical. We look for oracle manipulation, MEV extraction paths, share math that drifts under load, and governance attacks that make stealing profitable. Your invariants matter more than your linter warnings.
RWA (Real World Assets): The risk lives at the seams: where on-chain state meets off-chain custody, where attestations can be replayed, where governance can silently rewrite backing assumptions. We audit the full trust chain—not just the contract code.
NFT projects: Minting rules, transfer restrictions, marketplace logic. The failure modes are specific: unauthorized duplication, ownership state desyncs, allowlist bypasses, royalty enforcement gaps. Standard reentrancy checks don't catch these.
GameFi economies: Its an economy with game mechanics on top. We test for reward loops, emission exploits, anti-farming bypasses, and marketplace manipulation. If there as a way to print infinite value through "normal gameplay," we will find it.
Wallet contracts: These fail catastrophically. One signature validation bug, one replay vector, one recovery logic mistake—and user funds are gone or permanently locked. We review authorization like an attacker: what can I sign once and replay forever?
Different projects need different attack scenarios. Flash loan testing makes sense for DeFi, not for NFTs. Signature replay matters for wallets, not for lending pools. We build our test harnesses around what actually threatens your specific project type.
The invariants change too. DeFi needs "no free money" guarantees. RWAs need "supply always matches backing." NFTs need "no unauthorized minting." GameFi needs "rewards stay bounded." Wallets need "only authorized signers execute." Generic audits miss this.
We use Foundry for testing across all types, but the test scenarios change completely. Medusa and Halmos for property checking when the invariants are clear. Certora for formal verification on critical paths. The tooling is less important than knowing what to test for.

Reports that match your risk profile. DeFi clients get profit-path findings with economic impact. RWA projects get trust-chain analysis. NFT teams get concrete duplication scenarios. GameFi gets economy collapse reproduction. Wallet projects get authorization failure proofs. Fixes that actually work for your project type. Not copy-paste remediation advice—specific guidance on restoring the invariants that matter for what you're building.
We test what breaks for your type of project, not what breaks in general.

Frequently Asked Questions
Because they fail differently. DeFi breaks when economic incentives are misaligned—oracle manipulation, MEV extraction, liquidity drains. NFTs break when ownership rules are inconsistent—unauthorized minting, transfer bypasses, supply cap violations. Same underlying tech, completely different attack surfaces.
Wallets fail catastrophically—one signature validation bug and all user funds are at risk. The threat model is "can I authorize the wrong signer?" not "can I extract value from a pool?" We test replay resistance, recovery hijacking, and batching logic. DeFi audits don't cover these.
GameFi exploits look like gameplay until they don't. We test for reward loops that print infinite value, multi-account farming with weak anti-abuse, and marketplace manipulation that breaks game economies. DeFi testing focuses on financial primitives; GameFi testing focuses on game mechanics that happen to have financial consequences.
DeFi audits are expensive because:
Complex math (share calculations, AMM pricing)
External dependencies (oracles, other protocols)
Economic attack modeling required
NFT audits cost less because:
Simpler logic (minting, transfers, ownership)
Fewer external integrations
More standardized patterns
Wallet audits are high-stakes because:
Catastrophic failure mode (all user funds)
Signature validation is subtle
Recovery mechanisms are complex
RWA audits include extra work:
Off-chain trust verification
Attestation and custody review
Compliance considerations
nSLOC is a starting point, but:
2K lines of simple ERC-20 logic: 1 week
2K lines of custom AMM math with oracle dependencies: 3 weeks
Complexity = external calls + novel mechanisms + economic attack surface. We price based on actual risk, not just line count.
Supply and minting:
Allowlist bypasses through signature replay
Supply cap violations via reentrancy
Admin mint functions without proper access control
Transfer and ownership:
Approval bugs that allow unauthorized transfers
Batch transfer logic that desyncs ownership state
Unsafe callbacks that enable reentrancy drains
Marketplace integration:
Royalty bypass patterns
Price manipulation in listing logic
Reentrancy in exchange flows
Choose type-specific if:
You're building something novel within a category (custom AMM for DeFi, play-to-earn mechanics for GameFi)
Your project has high TVL or user funds at stake
You're integrating with complex external systems (bridges, oracles, other protocols)
Standard audit works if:
You're using well-tested patterns (OpenZeppelin ERC-20, standard NFT)
Your contract is simple and self-contained
You're not handling significant value
A blockchain security audit firm with the goal of making the Web3 space more secure through innovative and effective solutions.