GapLeak: When Privacy-Preserving LLM-based Web Agents Become a Privacy Risk
Large Language Model (LLM)-based web agents rely on data minimization and least privilege to protect user privacy. We show that these principles are necessary but insufficient: they constrain authorized sources, but not higher-sensitivity sources that remain reachable. A website can request a field unavailable from authorized sources, creating an information gap. To complete the task, a goal-driven agent may escalate to reachable but unauthorized sources and disclose accumulated sensitive information. We formalize this vulnerability as GapLeak, where an adversary exploits a legitimate web form without prompt injection or other malicious instructions. We evaluate GapLeak on three web-agent frameworks and six LLMs from Anthropic, Google, and OpenAI across recruitment, e-commerce, travel, and healthcare. GapLeak induces an average unauthorized disclosure of approximately 37%. The information gap is the primary driver: eliminating it reduces disclosure by more than 50%. Existing instruction-level defenses based on least privilege partially mitigate the vulnerability but often reduce task success. We propose submission-time verification, which filters unauthorized content before form submission, eliminating over 96% of unauthorized disclosures while preserving task success.
Authors
- Mohammad Zulkernine (ORCID: https://orcid.org/0000-0003-1697-4101)
- Jianbing Ni (ORCID: https://orcid.org/0000-0002-5639-0883)
- Tiancong Cao (ORCID: https://orcid.org/0009-0003-2958-2235)
- Rongxing Lu (ORCID: https://orcid.org/0000-0001-5720-0941)
- Chaklam Cheong
- Zelin Zhang
- Qi Li
Publication Details
- Journal
- arXiv (Cornell University)
- Published
- 2026-11-17
- DOI
- https://doi.org/10.48550/arxiv.2611.22223
- Primary Topic
- Web Application Security Vulnerabilities
- Type
- preprint