add-archon-model

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and analyze HuggingFace model files (e.g., "Read the HuggingFace model's source code" and "Fetch and analyze the model's HuggingFace configuration and modeling files" via AutoConfig.from_pretrained) — public, user-contributed content that the agent must interpret and that directly determines code-generation and behavior, enabling indirect prompt injection.