ai-avatar-video

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). The bundle includes an explicit instruction to pipe and execute a remote shell script (curl ... | sh from https://cli.inference.sh) and hosts user-uploaded/cloud files, a common high-risk pattern for malware distribution even if other links are just docs/media.

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to fetch and process arbitrary external media URLs (e.g., the example commands using "image_url", "audio_url", and "video_url" like "https://portrait.jpg", "https://speech.mp3", "https://video.mp4") so untrusted, user-provided web content is ingested and directly drives subsequent processing steps (transcription, TTS, lipsync), enabling indirect prompt-injection influence.

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

• Potentially malicious external URL detected (high risk: 1.00). The Quick Start includes a runtime command that pipes remote code to the shell ("curl -fsSL https://cli.inference.sh | sh"), which fetches and executes external code from https://cli.inference.sh and is presented as a required installation step.