vue-ai-assistant

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

• Third-party content exposure detected (high risk: 0.85). The required workflow fetches https://alexop.dev/llms.txt and then has runSubagent read linked pages from alexop.dev (public web content authored by Alex/outsider relative to the operating user), which becomes readable text in the agent’s LLM context via the subagent’s research step.

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

• Potentially malicious external URL detected (high risk: 1.00). This skill explicitly fetches https://alexop.dev/llms.txt at runtime and uses that fetched content to direct the agent's prompt selection and the runSubagent research workflow, so the URL is a required external dependency that directly controls agent instructions.