electrobun-platform

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly shows runtime webviews loading arbitrary URLs (e.g., the example) and the Security Checklist warns about "sandbox: true on any BrowserView loading untrusted URLs" and navigation rules, indicating the skill's runtime can ingest and act on untrusted third‑party web content which could carry indirect prompt-injection vectors.