andrej-karpathy-perspective

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md "Step 2: Karpathy式研究" explicitly mandates using tools (WebSearch etc.) to fetch and read public third‑party sources—e.g., GitHub, X/Twitter posts, blogs, news and forum feedback—which the agent must interpret to drive its research output and subsequent decisions, thus exposing it to untrusted user‑generated content that could carry indirect prompt injections.