Knowledge files have been widely used in large language model (LLM) agents, such as GPTs, to improve response quality. However, concerns about the potential leakage of knowledge files have grown significantly. Existing studies demonstrate that adversarial prompts can induce GPTs to leak knowledge file content. Yet, it remains uncertain whether additional leakage vectors exist, particularly given the complex data flows across clients, servers, and databases in GPTs. In this paper, we present a comprehensive risk assessment of knowledge file leakage, leveraging a novel workflow inspired by Data Security Posture Management (DSPM). Through the analysis of 651,022 GPT metadata, 11,820 flows, and 1,466 responses, we identify five leakage vectors: metadata, GPT initialization, retrieval, sandboxed execution environments, and prompts. These vectors enable adversaries to extract sensitive knowledge file data such as titles, content, types, and sizes. Notably, the activation of the built-in tool Code Interpreter leads to a privilege escalation vulnerability, enabling adversaries to directly download original knowledge files with a 95.95% success rate. Further analysis reveals that 28.80% of leaked files are copyrighted, including digital copies from major publishers and internal materials from a listed company. In the end, we provide actionable solutions for GPT builders and platform providers to secure the GPT data supply chain.

One of the reasons for consumers to buy into Microsoft's Copilot Pro, Custom GPT Builder, is quietly being killed off. Microsoft said that GPT Builder is being retired, and if you have bought into it, you'll have to remove all of your data by July 14. Unfortunately, Microsoft doesn't consider the ability to build GPT models a consumer priority, the company wrote in a support document. Microsoft Copilot is a "neutral," Microsoft-specific version of Copilot. You can think of GPT Builder as a way to personalize Copilot in much the same way that you can personalize your PC: GPT Builder allowed you to train a custom Copilot on your own data, with your own style.