AI code security free to start, for smart contracts and ZK circuits

Scout reviews on the platform, AuditHub MCP works inside your own coding agent, and formal methods tools test the findings, so you act on evidence, not on a model's guess.

TRUSTED BY

AI security code review backed by evidence

Scout reviews your code on the AuditHub platform, then tests each finding with the right security tool, so every finding comes with its evidence.

Meet Scout
Basic Code Review run: 11 findings (2 critical, 1 high, 3 medium, 4 low, 1 info) and the top three findings on ShieldedPool

Basic code reviewer

Reviews your code the way a human reviewer would, even on code that doesn't compile yet.

// BEST FOR EARLY BRANCHES
Tool Assisted Review: the Scout finding 'First depositor can inflate the share price' next to the OrCa counterexample, the exact transaction sequence that breaks the share price invariant

Tool assisted code reviewer

Plans and runs Vanguard, OrCa, ZK Vanguard and Picus against its threat model. Each finding comes with a location, severity, confidence score and the tool output, including counterexamples.

// BEST FOR AUDITS AND RELEASES

AI code security that remembers your protocol

Scout keeps the threat model and your corrections between runs, so your credits go to new code, not to relearning the old.

A coding agent is asked to add input validation and check it with AuditHub before commit. It calls AuditHub MCP, runs OrCa and attaches the result: a counterexample for a zero amount withdrawal on ShieldedPool.

Secure the code your coding agent writes

AuditHub MCP lets your coding agent call the security tools during the session, so the code it writes is checked before you commit it.

Keep your coding agent.

AuditHub MCP works inside the agent you already use.

Read what it sends.

The server is open source, so you can check every call it makes.

Install AuditHub MCP

Start AI code security for free

Every security tool is available from the Free plan.

Start for Free Now
  1. Sign up

    Sign up with GitHub or Google.

  2. Connect a repository

    Or install AuditHub MCP in your coding agent.

  3. Get your first findings

    With the evidence behind them.

AI code security without giving up your code

You stay in control of your code, your spend and your results, whichever route you use.

Read our Trust page
  • Read what the MCP server sends

    The server is open source.

  • Cap every run

    Scout stops at the credit budget you set and keeps every finding it made.

  • Export your findings

    Download report.md and findings.json from each run.

  • Keep your code out of training

    Your code never trains a model, ours or the model providers'.

AI code security FAQ

Everything you need to know about Scout, how it reviews your code, uses security tools, manages credits, and fits into your security workflow.

What is AI code security?

AI agents review your code for vulnerabilities. On AuditHub, security tools built on formal methods check what the agents find, so you act on evidence, not on a model's guess.

Can I start for free?

Yes. Every security tool is available from the Free plan. Plans and credits are detailed on the Pricing page.

Should I use Scout or AuditHub MCP?

Use AuditHub MCP to check code while your coding agent writes it. Use Scout for a full review of your protocol before an audit or a release. You can use both.

Does it work on ZK circuits?

Yes. ZK Vanguard catches likely underconstrained signals; Picus proves their absence.

Does AI code security replace a smart contract audit?

No. AuditHub checks your code as you build, so your audit starts from a higher baseline. The audit still gives you a signed expert review.

Still have questions? Contact us

Start your first AI security review

Run Scout on your protocol, or bring the security tools into your coding agent.