telnyx-ai-inference-java

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

• Third-party content exposure detected (high risk: 0.80). The skill includes an "Embed URL content" endpoint (POST /ai/embeddings/url) that explicitly crawls and ingests website content (including child pages) from arbitrary URLs into a storage bucket and the docs also show downstream similarity-search and embedding workflows that would cause the agent to read and act on that untrusted, user-provided web content, enabling indirect prompt injection.