docyrus-ui-react-components

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime workflow is to fetch llms.txt pages from https://ui.docy.app (public web content) via WebFetch/curl, and those fetched pages are free-form text that can contain prompt-injection payloads; this is an outsider source path: “public web content fetched at runtime → LLM ingests fetched llms.txt prose.”

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

• Potentially malicious external URL detected (high risk: 1.00). This skill explicitly instructs the agent to fetch remote llms.txt documentation at runtime (e.g. https://ui.docy.app/llms-web.txt and per-component pages like https://ui.docy.app/docs/web/components/data-grid/llms.txt), and those fetched files are the required source-of-truth that will directly control the agent's prompts/instructions for generating code.