rozenite-agent

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

• Third-party content exposure detected (high risk: 0.80). The skill instructs the agent to discover and read runtime plugin-provided domain metadata and live tool/schema lists (via "rozenite agent domains --session " and subsequent " tools" / "schema" calls) and to fetch request/response bodies from the app's network domain (e.g., getResponseBody), which are untrusted third-party runtime contents that can drive what tools/commands the agent uses, so malicious content could indirectly inject instructions.