qwen-image-2-pro

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill is a standard image-generation wrapper, but its first instruction installs a CLI from a different publisher (belt-sh/cli) than the inference-sh org used everywhere else, which is an unexplained supply-chain divergence rather than clear malicious intent.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (medium risk: 0.30). The skill's operational install instruction, npx skills add belt-sh/cli (line 7), pulls the belt CLI from a GitHub organization that differs from the documented platform's inference-sh org, which is a concrete but not conclusive signal of a decoy source.