EUVD-2026-36715
GHSA-9jq6-3jwj-jrfh
Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessCraft PDF with malicious JavaScript
-
DeliveryDeliver via email or web download
-
ExploitVictim opens PDF in Foxit AI
-
ExecutionSandbox fails to intercept dangerous interface
-
PersistEmbedded script fetches remote payload
-
ImpactArbitrary code executes as user
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A08:2025 - Software or Data Integrity Failures
ASVS 5.0
V15 - Secure Coding and Architecture Trusted source inclusion only
|
| Exploitation Conditions | Exploitation requires a victim to open an attacker-crafted PDF inside Foxit AI with JavaScript processing enabled (the default), and the host must be able to reach the attacker's remote script URL for the second stage to load. Limiting factors are the required user interaction (UI:R), local attack vector (AV:L) meaning the malicious file must reach the endpoint via a separate delivery channel, and outbound network connectivity from the Foxit AI process; environments that disable PDF JavaScript or strictly egress-filter document readers materially reduce realistic risk. |
| Impact | The CVSS 3.1 vector (AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H) describes local attack vector requiring user interaction, but the description states the malicious PDF loads remote scripts - meaning the initial delivery is plausibly via email or web download even though execution context is local to the victim. EPSS probability is very low (0.13%, percentile 3%) and SSVC marks Exploitation as 'none' and Automatable as 'no', consistent with no public exploit identified at time of analysis. Technical impact is rated total (full C/I/A loss), so a successful lure against a single user yields complete system compromise; risk should be treated as moderate-to-high for endpoints that routinely open untrusted PDFs and lower for hardened, restricted-document environments. |
| Exploit ScenarioDerived | An attacker crafts a PDF containing JavaScript that invokes the unmediated sandbox interfaces to fetch a second-stage payload from an attacker-controlled server, then emails or links the document to a target. When the victim opens the file in Foxit AI, the embedded script loads and executes the remote code with the privileges of the user. No public exploit identified at time of analysis, but the description is sufficiently detailed that proof-of-concept development is feasible. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv3 – Jun 16, 2026
Arbitrary code execution in Foxit AI (versions before 2026-06-15) occurs when the application processes JavaScript embedded in a PDF inside its sandbox but fails to intercept dangerous interfaces, permitting remote script loading. A user must open a malicious PDF for exploitation to succeed, and no public exploit identified at time of analysis with EPSS at only 0.13%. SSVC reports total technical impact but no observed exploitation.
Earlier versions of this analysis (2)
Arbitrary code execution in Foxit AI allows attackers to escape the embedded PDF JavaScript sandbox by invoking unintercepted dangerous interfaces, which can load remote scripts and execute arbitrary code on the victim host. Triggering requires a user to open a malicious PDF in Foxit AI; no public exploit identified at time of analysis, and EPSS exploitation probability is low (0.13%).
Arbitrary code execution in Foxit AI occurs when the application processes JavaScript embedded in a PDF document, because the sandbox fails to block certain dangerous interfaces that permit loading of remote scripts. A victim who opens a malicious PDF can have attacker-controlled code executed with scope change beyond the sandbox boundary, and no public exploit identified at time of analysis.
RemediationAI
Patch available per vendor advisory: upgrade Foxit AI to a build dated 2026-06-15 or later per the Foxit security bulletin at https://www.foxit.com/support/security-bulletins.html. Until upgrade is possible, disable JavaScript execution in PDF preferences (this breaks legitimate interactive forms and dynamic content), block opening of PDFs from untrusted sources at the email gateway and web proxy, and restrict outbound network access from the Foxit AI process so the sandbox cannot reach attacker-controlled script hosts - note that endpoint egress filtering may break other legitimate remote-content features. User-awareness reminders to avoid opening unsolicited PDFs are a partial compensating control given the UI:R requirement.
Affected VersionsENISA
Affects your stack
Find out whether the stacks and servers you run are on an affected version.
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Foxit AI integrates JavaScript processing for PDF documents, executing scripts inside a sandbox intended to constrain access to host resources. CWE-829 (Inclusion of Functionality from Untrusted Control Sphere) identifies the root cause: the sandbox does not fully mediate certain JavaScript interfaces, allowing the embedded script to fetch and execute remote payloads from attacker-controlled origins. Affected products per CPE are cpe:2.3:a:foxit_software_inc.:foxit_ai:*:*:*:*:*:*:*:*, indicating all Foxit AI builds prior to the fix date are in scope.
Know when Foxit Ai gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
Local file open with user interaction (AV:L, UI:R), no privileges needed (PR:N), low complexity once PDF is opened, and full code execution yields C:H/I:H/A:H.
Primary rating from Vendor (Foxit).
CVSS VectorNVD
Lifecycle Timeline
3DescriptionNVD
When the application executes the JavaScript script embedded in the PDF within the sandbox, it fails to intercept some dangerous interfaces, which allows remote scripts to be loaded, resulting in arbitrary code execution.
Executive ViewAI
Foxit AI users are at risk of system compromise through opening malicious PDF files, potentially exposing the organization to data theft and operational disruption. While currently no active attacks have been observed, all employees using Foxit AI for document processing remain vulnerable until the software vendor releases a security update.
Until vendor patch available: (1) Disable JavaScript execution in Foxit AI preferences across all installations; (2) Restrict PDF opening from external or untrusted email sources via endpoint controls; (3) Deploy alternative PDF readers (Adobe Acrobat Reader) for sensitive documents; (4) Establish user security awareness program focusing on PDF-based threats; (5) Create escalation process for suspicious or unexpected PDF receipt.
Threat Intelligence Overview
NVD
CVE.org
EPSS
Vulnrichment
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.
Exploit-DB