Claude Pirate: Abusing Anthropic's File API For Data Exfiltration

<p>Recently, Anthropic added the capability for Claude&rsquo;s Code Interpreter to perform network requests. This is obviously very dangerous as we will see in this post.</p> <p>At a high level, this post is about a data exfiltration attack chain, where an adversary (either the model or third-party attacker via indirect prompt injection) can exfiltrate data the user has access to.</p> <p><a href="