Command every trade. Protect every key.

Give every AI agent isolated access, deterministic limits, and proof for every transaction.

Built for agent execution. Governed by human intent.

AliasIsolateLimitTrace

Your wallet should not become the agent's identity.

VeilGuard separates what an agent needs to execute from what it never needs to see.

Interactive policy demoNOVA-03
Ready for simulated policy evaluation.
Agent aliasvg:nova:7E4F

The primary address stays out of the agent interface.

Simulated policy run
0.80 ETHUSDC
Limit orderApproved venue
Daily budgetWaiting
Asset allowlistWaiting
Approved venueWaiting
Human approvalWaiting
Demo receiptVG-6AE291

A simulated receipt appears after approval.

Ready for simulated policy evaluation.

Control every move an agent makes.

Privacy is useful only when permissions stay explicit and enforceable.

Revocable aliases
vg:nova:7E4F

Keep the primary address out of the interface.

Agents use a replaceable public alias while authorized records preserve the real account relationship.

Four physical gates inside a protected graphite enclosure
Isolated accounts

One agent. One boundary.

Each agent receives its own account, permissions, budget, and kill switch.

Deterministic limits
Asset allowlist ETH, WETH, USDCApproval threshold Required

Policy wins over the prompt.

Asset rules, approved venues, spending caps, and human approvals execute before every trade.

Integrity receipts

Every decision leaves a digest.

Requests, policy results, approvals, and human-signed testnet outcomes are bound to one integrity-checked record.

DigestVG-8C41F2Testnet verified

From intent to verified policy outcome.

Natural language proposes the action. Deterministic policy decides whether it can happen.

Connect

Assign a dedicated account to an AI agent.

Scope

Set assets, venues, limits, and approval rules.

Execute

Let the human wallet sign the exact policy-bound Base Sepolia request.

Verify

Review a human-only chain receipt or an explicit non-broadcast demo digest.

Clear boundaries

Private where it should be. Accountable where it must be.

Sensitive addresses stay out of AI-facing interfaces while authenticated human records remain complete and verifiable.

VeilGuard protects

  • Public interface address fields
  • Agent access to primary accounts
  • Budgets, assets, venues, and approvals
  • Authorized audit records

VeilGuard never does

  • Mix funds or conceal transaction origins
  • Alter or remove blockchain records
  • Bypass KYC, AML, sanctions, or reporting
  • Promise returns or provide investment advice

Privacy controls do not make transactions anonymous or remove legal obligations.

Give every AI agent a safer way to trade.

Start with isolated access, explicit limits, and proof for every action.

Create protected agent