Services

From the first sketch to life on mainnet.

Eight services, arranged by when a team usually needs them. Use any one on its own, or several together: each is scoped to your code and quoted before you commit.

Designing

Before the code is written

Security consulting

A senior researcher works through your design with you while it can still change: the architecture, the roles and privileges, how money moves, and what you are trusting. A flaw is far cheaper to fix on a whiteboard than in deployed code.

Ask about consulting

What it covers

  • A design review and a threat model you sign off
  • Roles, admin keys and upgrade paths
  • Trust in oracles, bridges and other protocols
  • The invariants that must always hold
  • Planning for key custody and launch

A good fit when

You are designing a protocol, planning a large change, or want a second opinion before committing to an architecture.

Building

While the code changes

Code scans

Static analysis on every release, using established tools and our own rules, merged into one list. A person confirms each lead before it is reported, so you see what is new since the last scan without the noise.

Ask about code scans

What it covers

  • Runs on each release, before deployment and after upgrades
  • Results from every tool merged, duplicates removed
  • Every lead confirmed by a person before it reaches you
  • What is new and what has been fixed since the last run

A good fit when

Your code changes often and you want problems caught as they are written, not months later.

Code review

A focused manual review of one change: a pull request, an upgrade or a new module, read against the version that is already deployed.

Ask about code review

What it covers

  • The diff reviewed against what is deployed
  • The migration plan checked
  • Findings scored with OVSS, or CVSS for off-chain code
  • A fixed quote for the change

A good fit when

You have been audited before and are shipping an upgrade or a new feature.

Launching

Before it goes live

Smart contract audits

A full review of your on-chain code by a senior researcher, in two tiers. Core is static analysis plus a manual review of your flattened code. Fork adds working exploits written against a fork of the live network.

Request an Audit

What it covers

  • Every finding backed by proof: a local test, a traced code path or a fork exploit
  • Severity scored with OVSS, which you can recompute yourself
  • A re-review of your fixes
  • A letter grade, A to F, for the audited commit
  • Solidity, Rust for Solana and CosmWasm, Move for Aptos and Sui, Go for Cosmos SDK

A good fit when

You are heading for a code freeze before a launch or a major upgrade.

Dapp security

Plenty of protocol losses never touch a contract bug. We test everything around the contracts the way an attacker would: the frontend, wallet and signing flows, APIs, bots and keepers, and the infrastructure behind them.

Ask about dapp security

What it covers

  • The frontend and its supply chain: dependencies, third-party scripts, hosting
  • Wallet connections and what users are asked to sign
  • APIs, bots, keepers and other off-chain services
  • The traditional web vulnerability classes
  • Deployment keys and pipelines

A good fit when

Your users reach your protocol through an app you run.

AI/ML security

For products where an AI model reads user input, calls tools or moves funds. We test how it can be misled and what it can be made to do.

Ask about AI/ML security

What it covers

  • Prompt injection, typed directly or hidden in content the model reads
  • Agents with wallet, key or tool access, and the limits placed on them
  • Data leaking through prompts, context and logs
  • The supply chain behind the model and its data
  • Guardrails tested by trying to get past them

A good fit when

You are shipping an AI agent, assistant or trading bot that touches user data or funds.

Live

Once real money is in it

Red team engagements

An agreed attack on your whole organisation, run the way a determined adversary would: the protocol, the apps and infrastructure, and the people and processes that hold the keys. Led by a researcher with over eight years of Fortune 500 penetration testing.

Ask about red teaming

What it covers

  • Scope and rules of engagement agreed in writing before anything starts
  • Attack paths that chain web, infrastructure and on-chain weaknesses
  • Key custody, signing and deployment processes
  • Social engineering, only if you put it in scope
  • A report of what got through and how to close it

A good fit when

You are live and want to know how your defences hold up against a real attacker, not only how your code reads.

Get Shielded

Ongoing protection after the audit. Your code is scanned on every release, each upgrade is reviewed before it ships, and your grade stays current instead of going out of date.

Get Shielded

What it covers

  • Code scans on every release
  • A review of each upgrade before it is deployed
  • Your grade re-checked after each change
  • A heads-up when an exploit hits a protocol like yours
  • A direct line to a senior researcher

A good fit when

You have launched and ship changes regularly.

Not sure where to start?

Tell us what you are building and where it is up to. We will suggest what fits.

Contact Us