Agent 自动支付场景的信任模型存在系统性问题,影响面延伸到 LLM 工具链。
ai_agent报告绕过
RT @v12sec: V12 found 25 vulnerabilities in the x402 protocol.
Critical findings included:
- Requests approved without any payment
- Maliciously high requests being paid by default
- Session takeover from relayed login signatures
Here's how. 🧵 https://t.co/0jIv7VpET7