emblem-ai-react

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

• Third-party content exposure detected (high risk: 0.80). The skill's migratefun hooks and project APIs (references/migratefun-react.md) explicitly fetch public project and token metadata from {baseUrl}/api/migrate-fun/ (default https://emblemvault.dev) and token metadata URIs (project.oldTokenMeta.uri) — untrusted, potentially user-generated third‑party content that the agent is expected to read and use in UI and decision-making (e.g., mint info, images, project data) at runtime.

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

• Potentially malicious external URL detected (high risk: 0.90). The HustleProvider example configures hustleApiUrl="https://agenthustle.ai", which is contacted at runtime by the chat/provider (see systemPrompt and skipServerPrompt) and can supply server-side prompts that directly control the assistant's behavior, so this external URL is a runtime dependency that can control prompts.