ace-step

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

• Third-party content exposure detected (low risk: 0.10). The workflow fetches model documentation from a vendor URL (https://www.pixazo.ai/models/ace-step.md), which falls under low risk as an actively fetched, trusted provider documentation source.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent at runtime to fetch external instructions and reference documentation from https://www.pixazo.ai/models/ace-step.md to control its behavior and parameter handling.