winmd-api-search

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to read and act on a local JSON cache that is generated from third‑party NuGet packages and project-output WinMD files (see "Cache Setup" / "Read the JSON to choose the right API" in SKILL.md and the Search-WinMd/Invoke-WinMdQuery.ps1 logic which outputs JSON file paths for the agent to inspect), so untrusted third‑party package metadata can be ingested and materially influence subsequent API selection and code-generation decisions.