Agent Payment Security Audit

Can a compromised agent move money outside policy?

Nine questions about your agent's payment setup, scored on eight dimensions. Anything you leave out comes back as Unknown with the exact question to ask. The blast radius is in your own numbers. This is a diagnostic, not a sales tool.

npx @olurabian/audit

01 What the agent buys

What does the agent buy, and how often? API credits, compute, data, vendors, on-chain.

02 What the runtime can reach

What can the agent's runtime reach? List every tool, MCP server, SDK, and key in the agent's process.

03 Where the credential lives

Where does the payment credential live? In the agent's process or prompt or memory, or behind a separate service or signer.

04 Who executes the payment

Who executes the payment? The agent calls the rail itself, or it submits an intent to something that executes.

05 What a grant is bound to

When a spend is approved, is it bound to a specific payee and amount, or can the agent supply any in-policy value?

06 Limits and where they bite

What limits exist and where are they enforced? Per-action, daily, per-vendor. Checked before the spend or only at settlement.

07 Human approval

Is there human approval for large spends? None, in-band where the agent decides, or out of band where a person approves the exact spend.

08 The record

Is there a record of every decision and the amount actually settled, and can it be tampered with?

09 Drift

How often does the agent's tool or dependency set change, and is the money-path re-checked when it does?

Runs in your browser. Nothing you type leaves this page.