CVE-2025-47783 label-studio vulnerable to Cross-Site Scripting (Reflected) via the label_config parameter.
Attackers can inject malicious scripts via the label_config parameter in Label Studio, leading to data theft and session hijacking.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Over the internet
- Victim action needed
- A user has to click or open something
- Fix
- Available
What is CVE-2025-47783?
CVE-2025-47783 is a high-severity cross-site scripting vulnerability in HumanSignal label-studio, published on 14 May 2025. An attacker can exploit it over the internet without logging in, and only if a user opens a file or follows a link.
Official description
Label Studio is a multi-type data labeling and annotation tool. A vulnerability in versions prior to 1.18.0 allows an attacker to inject a malicious script into the context of a web page, which can lead to data theft, session hijacking, unauthorized actions on behalf of the user, and other attacks. The vulnerability is reproducible when sending a properly formatted request to the `POST /projects/upload-example/` endpoint. In the source code, the vulnerability is located at `label_studio/projects/views.py`. Version 1.18.0 contains a patch for the issue.
Weakness: CWE-79
Which versions are affected by CVE-2025-47783?
3 open source packages ship an affected version, including label-studio. It affects 1 product.
| Package | Affected | Fixed in |
|---|---|---|
label-studio PyPI | All versions | 1.18.0 |
label-studio PyPI | All versions | 1.18.0 |
label-studio PyPI | All versions | 1.18.0 |
Products include Humansignal Label Studio.
How do I fix CVE-2025-47783?
Upgrade label-studio to 1.18.0 or later.
Escape output for the context it lands in and set a strict Content Security Policy. Upgrading the affected component is the real fix. More about Cross-site scripting.
Is CVE-2025-47783 being exploited?
It is not confirmed as exploited in the wild, but public exploit code for CVE-2025-47783 exists.
Public exploit code: 1 Nuclei template.
Other vulnerabilities in HumanSignal label-studio
- CVE-2023-43791 Label Studio has Hardcoded Django `SECRET_KEY` that can be Abused to Forge Session Tokens
- CVE-2026-22033 Label Studio vulnerable to full account takeover by chaining Stored XSS + IDOR in User Profile via custom_hotkeys field
- CVE-2025-25297 Label Studio allows Server-Side Request Forgery in the S3 Storage Endpoint
- CVE-2023-47117 Object Relational Mapper Leak Vulnerability in Filtering Task in Label Studio
- CVE-2023-47115 Label Studio XSS Vulnerability on Avatar Upload
Sources
Also known as GHSA-8jhr-wpcm-hh4h, PYSEC-2025-124.