model-compare

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

• Third-party content exposure detected (low risk: 0.10). The skill workflow involves running an AI model comparison script that fetches responses from provider APIs and uses a tool-calling loop with web_search and web_extract on arbitrary web pages. However, the outsider-authored text ingested from the web via search/extraction is not fed into an automated feedback loop or agent control channel without selection, and the primary required workflow is model evaluation/comparison driven directly by user prompts.