Launch offer20,000 credits free,a $200 value. Start for freeRun Scout on your code now.

Scout: The AI code reviewer built on auditors’ findings

Point Scout at your codebase. Scout combines several AI models with our skills and threat models to find bugs, and ranks every finding by severity, even on code that doesn’t compile yet.

How Scout works

Step 1 of 4

Scout reviews your code.

Point Scout at your codebase. It maps the architecture and likely threats, then several AI models, our skills and threat models review it the way a reviewer would, even if it doesn't compile yet.

20,000

credits, worth $200

Free plan: what you get

  • The Basic Code Reviewer and in-house tools
  • A credit budget you set for each review
  • No credit card required

Basic and Tool Assisted Code Reviewer

Two agents, your choice: a first read on any branch, or tool-backed findings before an audit or a release.

Basic Code Review run: 11 findings (2 critical, 1 high, 3 medium, 4 low, 1 info), 2,330 credits consumed in 41m 8s, 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
Coming soon
Tool Assisted Review: the Scout finding "First depositor can inflate the share price" (critical, high confidence, Vault.sol:142) next to the OrCa counterexample: deposit 1, transfer 1000 to the vault, then a victim deposits 500 and receives 0 shares

Tool Assisted Code Reviewer

Plans and runs several security tools against its threat model, including open-source tools and our own: Vanguard, OrCa, ZK Vanguard and Picus. Each finding comes with a location, severity, confidence score and the tool output, including counterexamples.

// Best for audits and releases

Scout vs a general AI reviewer

Scout runs on frontier models like other AI reviewers, and adds the memory, the tools and the budget a chat session lacks.

Criteria
Scout
General AI reviewer
Available now
Protocol understanding
ScoutBuilds a threat model around your protocolMaps its architecture and proposes key invariants.
General AI reviewerWorks from the context provided in the conversation
Review budget
ScoutA credit budget you control.Set a budget for each review. Scout stops at the limit and keeps the findings already made.
General AI reviewerUsage depends on the assistant and plan
Coming soon
Your feedback
ScoutComing soonRemembers your protocol between runs.Retains its threat model and your corrections for future reviews.
General AI reviewerFeedback primarily guides the current conversation
Findings grounded in
ScoutComing soonTests findings with specialized security tools.Uses static analysis, fuzzing, and ZK security tools where applicable, attaching evidence to findings.
General AI reviewerPrimarily model reasoning and provided context

Scout roadmap

Available now

Basic Code Reviewer

Point Scout at your codebase. Several AI models, our skills and threat models find bugs and rank them by severity.

Q4 2026

Persistent memory and feedback

Your corrections and the protocol model Scout builds carry from one run into the next.

Q4 2026

Open-source Rust tools

Open-source security tools for Rust smart contracts, available in AuditHub.

Q1 2027

Tool Assisted Code Reviewer

Several security tools, open-source and our own (Vanguard, OrCa, ZK Vanguard, Picus), run against the threat model, so findings come back with tool output behind them.

AI security review you stay in control of

Scout works within the credit budget you set and logs every task with why it ran, so you see exactly what it did with your code.

  • Set a credit budget

    Scout stops before going over and keeps every finding, even if the review didn’t finish.

  • Read why each task ran

    Every tool Scout runs is logged with its reasoning.

  • Choose who sees it

    Findings and Scout’s memory stay with the members of your AuditHub organization.

  • Remove a project

    Delete a project when you no longer need it. This action can’t be undone.

Questions before you start

Do I need a credit card to sign up?

No. The Free plan asks for no payment details, so nothing can be charged automatically. You get 20,000 credits, worth $200. When they run out, you choose: buy credits, upgrade, or stop.

What do I get on the Free plan?

20,000 credits, worth $200, the Basic Code Reviewer and in-house tools, with a credit budget you set for each review. Every plan is on the Pricing page.

How is Scout different from asking Claude or Codex to review my code?

Scout adds what a chat session lacks: our skills and threat models, and a credit budget you control. Tool-backed evidence and memory between runs are coming soon.

Is my code safe with Scout?

Your code never trains a model, and findings stay with your AuditHub organization.

How does the budget and credit system work?

You set a credit budget for each review. Scout stops before going over and keeps every finding made so far.

What can Scout review today?

Any codebase, with a specialization in smart contracts (Solidity/EVM) and ZK circuits.

What happens when I disagree with a finding?

Mark it as a false positive. Carrying that correction into the next run is coming soon.

See Scout on your codebase.