modal-deployment

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

• Third-party content exposure detected (high risk: 0.80). The skill's examples explicitly fetch and return content from arbitrary public URLs (for example the "async def fetch_urls(url: str)" example using aiohttp.ClientSession in the Concurrency/Scaling references), meaning the agent could ingest untrusted, user-provided web content that could carry indirect prompt injections.