stitch-vue-layui-components

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path fetches outsider-authored Stitch HTML from htmlCode.downloadUrl (a URL provided by the Stitch MCP response) via scripts/fetch-stitch.sh using curl, and that downloaded HTML becomes readable text that the agent parses into Vue code (public/remote content category: fetched at runtime from an arbitrary download URL).

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

• Potentially malicious external URL detected (high risk: 0.80). The skill runs scripts/fetch-stitch.sh (which does curl -L ... "$URL") at runtime to download the Stitch htmlCode.downloadUrl (e.g., Google Cloud Storage URLs returned by Stitch MCP), and that fetched HTML is injected into the conversion process so remote content can directly control the agent's generated code/output.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)