Stolen-Credential Account Takeover & Crypto Drain
This detects an AI agent being asked to use stolen banking credentials to break into a victim's account and move money to an attacker's cryptocurrency wallet. It covers a specific end-to-end fraud pattern: credential reuse, unauthorized login, and outbound transfer.
How the attack works
An attacker instructs an AI agent to log in using credentials that were not obtained legitimately, such as stolen bank usernames and passwords. The agent uses these to access the victim's account. It then initiates a transfer of funds out of that account to a cryptocurrency wallet controlled by the attacker. The rule includes a known fixed wallet address seen in this pattern, but the underlying behavior — stolen login followed by fund transfer — is the core signal, not just the wallet.
Netzilo detection
Netzilo reports this behaviour when it is observed.
- Signature ID
- 0f8e3c2d-19b5-4ad8-819c-03776906645b
- Severity
- Critical
Why it matters
Successful execution results in direct theft of money from a victim's financial account, with funds moved into cryptocurrency where they are hard to trace or recover.
What you can do
- →Require agents to use only credentials provisioned through an authorized, auditable identity system, never credentials supplied directly in a user prompt.
- →Block or flag any agent action that combines a login step with an outbound funds transfer to a new or unverified payee, especially crypto wallets.
- →Maintain and check against a blocklist of known fraud-associated wallet addresses, including 0x5d5acA289F2A9E481fa2aEaD3FA465880Df84354.
- →Log and review any agent session where credentials for a financial account are entered by the user rather than retrieved from a secure vault.
Known benign look-alikes
- Security/fraud teams describing how stolen-credential reuse is detected
- Educational explanation of account-takeover fraud
- Legitimate transfer of the user's own funds between their own accounts