CVE-2026-73032

PapersGPT for Zotero 0.6.1 RCE via Unsanitized LLM Response eval()

Published Aug 11, 2026 CVSS 4.0 · 9.4 Critical EPSS < 1% KEV · Not listed
Description
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.

Analysis

AI-generated

Technical impact

PapersGPT for Zotero 0.6.1 contains an unsanitized eval of JavaScript returned from an LLM endpoint, which permits an attacker to inject executable code. This flaw is identified as CWE-94 and results in remote code execution in Zotero’s chrome‑privileged context, enabling file read/write, process execution, and full access to all Zotero data.

Affected Systems

The vulnerability affects the PapersGPT for Zotero extension version 0.6.1 distributed by the papersgpt vendor. No other versions are listed as vulnerable in the CVE data.

Risk and Exploitability

The flaw carries a CVSS score of 9.4, classified as critical. The EPSS score is less than 1%, indicating a currently low exploitation probability, and the issue is not listed in CISA KEV. Attackers can exploit the vulnerability through prompt injection in PDFs, man‑in‑the‑middle interception of API requests, or by using a malicious custom LLM endpoint, all of which would enable code execution within Zotero’s privileged environment.

Remediation

No solution or workaround provided in the CVE record.

OpenCVE Recommended Actions

  • Update PapersGPT for Zotero to the latest patched version when available.
  • If a patch has not been released, uninstall or disable the PapersGPT for Zotero extension to remove the vulnerability surface.
  • Block or tightly control any custom LLM endpoint connections and monitor network traffic for suspicious API calls.

Generated by OpenCVE AI on August 12, 2026 at 20:09 UTC.

Tracking

Sign in to view the affected projects.

Advisories

No advisories available
History
Type Values Removed Values Added
First Time appeared Papersgpt
Papersgpt papersgpt-for-zotero
Vendors & Products Papersgpt
Papersgpt papersgpt-for-zotero
Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'poc', 'Technical Impact': 'total'}, 'version': '2.0.3'}

Type Values Removed Values Added
Description 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.
Title PapersGPT for Zotero 0.6.1 RCE via Unsanitized LLM Response eval()
Weaknesses CWE-94
References
Metrics cvssV3_1

{'score': 9.6, 'vector': 'CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H'}

cvssV4_0

{'score': 9.4, 'vector': 'CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:A/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H'}

Weaknesses
  • CWE-94

    Improper Control of Generation of Code ('Code Injection')

Sources

Status
PUBLISHED
Assigner
VulnCheck
Published
Updated
Nov 8, 2026, 2:08 PM
Reserved
Aug 10, 2026, 6:48 PM
Status
Deferred
Published
Aug 11, 2026, 8:18 PM
Modified
Oct 24, 2026, 3:10 PM