Voice AI Integration Engineer

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

• Third-party content exposure detected (medium risk: 0.65). SKILL.md:425-487 defines an LLM handoff payload that will include the runtime transcript text (content assembled from transcript["segments"]), and that transcript text originates from transcription of outsider-provided audio, which is attacker-controlled free-form content that can include prompt-injection strings.