aikey-benchmark

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

• Third-party content exposure detected (medium risk: 0.30). The workflow consumes outsider-authored free text from benchmark creator accounts fetched via API or manual extraction during social media channel teardowns.

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

• Potentially malicious external URL detected (medium risk: 0.30). An operational dependency on tikhub.io is defined in the configuration and guides for fetching benchmark data, but tikhub.io is a recognized third-party API service provider without first-party provenance, resulting in a medium provenance tier.