built-in-ai

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

• Third-party content exposure detected (low risk: 0.10). The fetch-idls script fetches index.bs specification files from arbitrary repositories on GitHub during updates, presenting a low prompt injection risk.

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

• Potentially malicious external URL detected (high risk: 0.70). The fetch-idls script fetches raw specification files from GitHub repositories (webmachinelearning) at build/update time, but the repositories are uncontrolled personal/community-level forks or non-official channels rather than established official publisher distribution mechanisms, leading to a weak or unknown operational dependency.