tooluniverse-image-analysis

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

• Third-party content exposure detected (low risk: 0.10). The workflow downloads data capsules from a Hugging Face dataset via HTTP requests, exposing the agent to potentially untrusted outsider-authored content.

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

• Potentially malicious external URL detected (high risk: 0.70). The test script references a Hugging Face dataset URL to download runtime data capsules, but the dataset is hosted on a third-party repository rather than an official project source.