lov-cli2anything

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The Skill contains guidance and runtime code that instructs users to load an unpacked browser extension with broad host permissions and to reuse existing logged-in browser sessions (including in-page fetches that read cookies/XSRF) — a security-weakening steering without clear malicious intent.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (medium risk: 0.30). The skill's discovery routine (discoverZenmuxData) constructs and fetches frontend chunk files from a hard-coded third-party CDN "https://cdn.marmot-cloud.com/page/tbox-router-home/" (e.g. https://cdn.marmot-cloud.com/page/tbox-router-home/4311e37b-async.d0071197.js), an operational external download controlled outside the project rather than a well-known vendor CDN.

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

• Third-party content exposure detected (low risk: 0.10). Required runtime workflow reads only a user-supplied/local discovery artifact file (e.g., --discovery artifacts/zenmux-discovered.json or --discovery <path>) and then generates/validates OpenAPI/SDK; it does not fetch arbitrary outsider-authored free text by default.