Cyber Resilience

CVE-2025-66481

RCE in Thinkinai Deepchat ≤ 0.5.1

Public PoCRCEXSS
Published
09 December 2025
Modified
11 December 2025
Patch / advisory
CVSS Score v3.1 9.6
Click a component to see what it means
Raw vectorCVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
EPSS Score 0.0056 44th percentile
Risk Priority 68 floored blend · peak EPSS

Summary

CVE-2025-66481 is a critical-severity Cross-site Scripting (CWE-79) vulnerability in Thinkinai Deepchat. Its CVSS base score is 9.6 (Critical).

Operationally, exploitation aligns with the MITRE ATT&CK technique Browser Session Hijacking (T1185); ranked at the 44th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.

This vulnerability is AI-related — categorised as LLM Application Platforms; in the Other ATLAS/OWASP Terms risk domain.

The strongest mitigations our analysis identified map to SA-11 (Developer Testing and Evaluation) and SI-10 (Information Input Validation) — see the control section below for these in your framework.

Deeper analysis AI-assisted summary

Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.

CVE-2025-66481 is a cross-site scripting (XSS) vulnerability affecting DeepChat, an open-source AI chat platform that supports cloud models and large language models (LLMs). Versions 0.5.1 and below are impacted due to improperly sanitized Mermaid content in the MermaidArtifact.vue component. A recent security patch for this component proves insufficient, as it can be bypassed using unquoted HTML attributes combined with HTML entity encoding, allowing attackers to evade the regex filter intended to strip dangerous attributes.

Remote unauthenticated attackers can exploit the vulnerability by tricking victims into interacting with maliciously crafted Mermaid content, requiring user interaction such as viewing or rendering the payload. Successful exploitation enables remote code execution (RCE) on the victim's machine via the Electron ipcRenderer interface. The issue carries a CVSS v3.1 base score of 9.6 (AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H) and is associated with CWE-79 (Improper Neutralization of Input During Web Page Generation), CWE-80 (Improper Neutralization of Script-Related HTML Tags), and CWE-94 (Improper Control of Generation of Code).

The GitHub security advisory (GHSA-h9f5-7hhf-fqm4) confirms there is no fix available at the time of publication, emphasizing the need for practitioners to avoid untrusted Mermaid content in affected DeepChat deployments until a patch is released.

OWASP Top 10 for Web (2025)

EU & UK References

Vulnerability Data

DeepChat is an open-source AI chat platform that supports cloud models and LLMs. Versions 0.5.1 and below are vulnerable to XSS attacks through improperly sanitized Mermaid content. The recent security patch for MermaidArtifact.vue is insufficient and can be bypassed using…

more

unquoted HTML attributes combined with HTML entity encoding. Remote Code Execution is possible on the victim's machine via the electron.ipcRenderer interface, bypassing the regex filter intended to strip dangerous attributes. There is no fix at time of publication.

CWE(s)

AI Security AnalysisAI

AI Category
LLM Application Platforms
Risk Domain
Other ATLAS/OWASP Terms
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: ai, llms

Related Threats

MITRE ATT&CK Enterprise Techniques

T1185 Browser Session Hijacking Collection
Adversaries may take advantage of security vulnerabilities and inherent functionality in browser software to change content, modify user-behaviors, and intercept information as part of various browser session hijacking techniques.
T1539 Steal Web Session Cookie Credential Access
An adversary may steal web application or service session cookies and use them to gain access to web applications or Internet services as an authenticated user without needing credentials.
T1659 Content Injection Initial Access
Adversaries may gain access and continuously communicate with victims by injecting malicious content into systems through online network traffic.
T1059 Command and Scripting Interpreter Execution
Adversaries may abuse command and script interpreters to execute commands, scripts, or binaries.
T1059.001 PowerShell Execution
Adversaries may abuse PowerShell commands and scripts for execution.
T1059.002 AppleScript Execution
Adversaries may abuse AppleScript for execution.
Derived from this CVE’s CWE(s) via the direct CWE→ATT&CK cross-walk.

CVEs Like This One

CVE-2023-40809Shared CWE-79, CWE-94
CVE-2023-1030Shared CWE-79, CWE-94
CVE-2023-4709Shared CWE-79, CWE-94
CVE-2023-45144Shared CWE-79, CWE-94
CVE-2024-25639Shared CWE-79, CWE-80
CVE-2023-0625Shared CWE-79, CWE-94
CVE-2023-1649Shared CWE-79
CVE-2026-45346Shared CWE-80
CVE-2025-66450Shared CWE-80
CVE-2026-41611Shared CWE-79, CWE-80

Affected Assets

thinkinai
deepchat
≤ 0.5.1

Mitigating Controls

Control response

Prevent
Stop it (NIST 800-53)

Detect
Catch it (NIST detect / respond)

Harden
Shrink the surface (DISA STIG)

Validate
Prove the fix (OWASP ASVS)
  • V1.1.2
  • V1.3.2
  • V1.2.1
  • V1.3.1

Mitigating Controls (NIST 800-53 r5) AI

Developer testing and evaluation can discover missing or incorrect input neutralization through targeted web-application tests.

Input validation directly enforces neutralization of untrusted data before it reaches web output generation.

Output filtering can catch or sanitize unneutralized script content before it is served to users.

Least privilege limits the damage an injected code fragment can perform once executed.

Requiring documented secure development standards and tools enforces use of safe code-generation APIs and escaping.

Secure engineering principles include mandatory output encoding and neutralization of HTML metacharacters to stop injection at the source.

Mitigating Controls (NIST CSF 2.0) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.

PR.PS-06 mostly match
prevents

Secure SDLC practices directly target introduction of XSS via coding standards/testing (mostly), yet the single broad outcome leaves many specific neutralization vectors unaddressed (partial).

PR.PS-02 partial match
prevents

Patching and EOL replacement can remediate known XSS instances in libraries or frameworks (partial) but do nothing to enforce input neutralization in application code (none).

PR.DS-10 none match
prevents

PR.DS-10 protects runtime data confidentiality/integrity but has no bearing on neutralizing externally influenced input during code generation, so neither direction shows any preventive effect.

Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.

finds

Secure-coding testing and automated code-analysis tools are applied to detect improper neutralization of script-related content during web-page generation.

prevents

Knowledge exchange on emerging attack techniques and patches reduces the likelihood that cross-site scripting flaws remain unaddressed in deployed applications.

prevents

Operational indicators of compromise for web-application attacks can be incorporated into WAF or input-filtering rules, lowering the likelihood that unsanitized data reaches the browser.

prevents

Requiring language-specific secure-coding standards and automated scanning during the SDLC catches missing output encoding or improper neutralization of untrusted data before the software reaches production.

prevents

Application security requirements explicitly call for neutralization of script-related HTML tags.

prevents

Secure-coding standards, SAST scans and removal of insecure code samples together eliminate the failure to neutralize script content that produces cross-site scripting flaws.

References