aiq-deploy

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

• Third-party content exposure detected (medium risk: 0.65). The runtime workflow can clone the public AI-Q repository from GitHub (git clone https://github.com/NVIDIA-AI-Blueprints/aiq in references/locate-or-clone.md), and the cloned repo’s files (e.g., docs/configs) can be read/ingested into the agent’s LLM context, which is outsider-authored free text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs cloning and using the remote repository https://github.com/NVIDIA-AI-Blueprints/aiq (e.g., "git clone https://github.com/NVIDIA-AI-Blueprints/aiq.git" and subsequent 실행 of scripts/docker-compose from that checkout), which fetches remote code at runtime that is then executed—so this is a runtime external dependency that can control behavior.