translate-polisher

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

• Third-party content exposure detected (low risk: 0.10). The workflow fetches external web pages via URL using curl -L and an intermediary service (r.jina.ai), which ingests outsider-authored free text at the agent's runtime via an active fetch triggered by user input.

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 URL content via curl -L from a third-party service (r.jina.ai), which acts as an operational runtime dependency providing markdown content used directly in the translation workflow.