minimax

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

• Third-party content exposure detected (low risk: 0.10). The workflow instructs the agent to fetch external model documentation from https://www.pixazo.ai/models/minimax.md, which is an actively searched URL rather than an unselected monitored input channel.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to fetch external instructions and reference data from https://www.pixazo.ai/models/minimax.md at runtime, which represents a weak/unknown third-party operational dependency.