andrej-karpathy-perspective

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

• Third-party content exposure detected (high risk: 1.00). The skill's required "Step 2: Karpathy-style research" explicitly mandates using WebSearch and external sources (e.g., GitHub, technical blogs, news outlets and "100+条X帖子"/Twitter-X) to fetch and interpret public, user-generated content that will directly influence the agent's analysis and subsequent actions.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)