wix-docs-base44

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

• Third-party content exposure detected (medium risk: 0.30). At runtime, the required workflow uses docs.search() and docs.fetchDoc()/readScratch()/mapTerms() to ingest outsider-provided free-text term (search query) and then LLM-processes the resulting markdown content fetched from Wix docs via the search workflow.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill posts arbitrary JS code to the remote code-execution endpoint https://mcp.wix.com/api/code-mode/search at runtime (via specQuery / methodsOf), which causes that external service to execute the provided code.