MediumVulnerabilityLLM-specific

CVE-2026-41318: AnythingLLM stored XSS through chart captions in chat history

Source: NVD(opens in a new tab)
Identifier: CVE-2026-41318
Published: April 24, 2026
Record updated: October 9, 2026

Known exploitation: Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
EPSS: 0.2%

Summary

AnythingLLM versions prior to 1.12.1 contain a stored DOM-level XSS flaw, tracked as CVE-2026-41318. The in-chat markdown renderer interpolates image `alt` text into an HTML attribute without encoding, and the `Chartable` component renders chart captions with no DOMPurify sanitization. An attacker who can influence LLM output, through indirect prompt injection in a shared workspace document or by creating a chart record in a multi-user workspace, can run script in other users' browsers when they open that conversation.

Mitigation

Patched in version 1.12.1.

Details

Scores and exploitation

CVSS score: 5.4(medium)
CVSS vector: CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:H/I:L/A:N
Attack vector: network
Attack complexity: high
Privileges required: low
User interaction: required
EPSS: 0.2%Estimated probability that this CVE is exploited in the next 30 days, from the FIRST Exploit Prediction Scoring System. Read on November 8, 2026.
Disclosure date: April 24, 2026

Weaknesses and techniques

Weakness (CWE): CWE-79
CWE-116
CWE-1336
Attack pattern (CAPEC): CAPEC-198
CAPEC-86
AI threat technique (MITRE ATLAS): AML.T0051

Classification

Attack type: Prompt injection,
Other
Attack sophistication: Moderate
AI component targeted: Plugin
Impact: Confidentiality,
Integrity
Labels: security

How this record was produced

Summaries, mitigations and labels on this page are generated automatically from the source. The linked source is authoritative. The source gives this advisory a description and no title, so the title above is a headline generated from the source text.

Original source: https://nvd.nist.gov/vuln/detail/CVE-2026-41318

First tracked: April 24, 2026 at 08:10 AM ET, from NVD/CVE Database

Classifier: claude-haiku-5-5, prompt v4

Classifier confidence: 90% (self-assessed)

Topics

Prompt injection and jailbreaks

Summary

Mitigation

Topics

Related items