CVE-2026-73032
CriticalCVSS 9.6Summary
PapersGPT for Zotero 0.6.1 contains a remote code execution vulnerability that allows attackers to execute arbitrary JavaScript by returning malicious code from an LLM endpoint that is passed unsanitized to window.eval() in views.ts. Attackers can exploit this through prompt injection in PDFs, MITM interception of API requests, or a malicious custom LLM endpoint to execute arbitrary code in Zotero's chrome-privileged context, enabling file read/write, process execution, and access to all Zotero data.
Risk Assessment
The risk includes full compromise of Zotero data and the user's system, as an attacker can access files, execute processes, and steal sensitive information.
Recommendation
It is recommended to update PapersGPT to a patched version, avoid using window.eval() with untrusted data, and verify the integrity of connections to LLM endpoints.
Original NVD description (English source)
PapersGPT for Zotero 0.6.1 contains a remote code execution vulnerability that allows attackers to execute arbitrary JavaScript by returning malicious code from an LLM endpoint that is passed unsanitized to window.eval() in views.ts. Attackers can exploit this through prompt injection in PDFs, MITM interception of API requests, or a malicious custom LLM endpoint to execute arbitrary code in Zotero's chrome-privileged context, enabling file read/write, process execution, and access to all Zotero data.