medusa-subscribers

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

• Third-party content exposure detected (high risk: 0.85). The required workflow explicitly instructs fetching live documentation from https://docs.medusajs.com/... and web-searching Medusa docs at runtime; those fetched pages are outsider-authored free text that would be ingested into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill repeatedly instructs the agent at runtime to "fetch live docs" (e.g., https://docs.medusajs.com/learn/fundamentals/events-and-subscribers) and relies on those fetched documents to shape implementation details and prompts, so the external URL is a runtime dependency that directly controls agent instructions.