huggingface-hub

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly shows the hf CLI fetching and interacting with public Hugging Face Hub content (e.g., "hf download REPO_ID", "hf datasets list" / "hf datasets sql", "hf discussions list"/"comment", and "hf extensions install"), which are untrusted, user-generated sources the agent would read or act on and thus could enable indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). The Quick Start includes an installation command that downloads and executes a remote script ("curl -LsSf https://hf.co/cli/install.sh | bash -s"), which runs remote code during setup and is a required dependency for the hf CLI.