V12 reports three critical flaws in x402 SDKs for AI agent payments

Security firm V12 published on 22 October 2026 its review of the x402 code, the open standard that lets software and AI agents pay for an API call inside the web request: 25 flaws, three of them critical. In the first, the default client set no spending limit, so a malicious API could get an agent that signs payments automatically to sign away its full balance of a supported token. In the second, the TypeScript Solana facilitator accepted a token approval, which moves no tokens, as a valid payment. In the third, the sign-in extension signed login messages for a website other than the one contacted. V12 reported the three critical and 13 high-severity findings to the x402 team on 12 August, and the critical ones were fixed by 20 August; nine medium and low findings followed on 9 September. V12 found the flaws by code review and local tests, and neither V12 nor the x402 Foundation, the Linux Foundation project that maintains the SDKs, has reported an attack using them. The Hacker News found that the foundation has published no security advisory for the three flaws and that npm audit and PyPI show no warning for the affected releases.

When

22 Oct 2026

ID

WTA-N-2026-0378

Where

International

Who

V12, x402 Foundation, Linux Foundation, Coinbase

Topics

AI agents, Cybersecurity

Spotted an error? Email [email protected].