# CVE-2026-72560: HumanSignal Label Studio server-side request forgery in URL import

Source: https://junglewise.ai/threats/cve-2026-72560-humansignal-label-studio-server-side-request-forgery-in-url  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-08-11. Updated: 2026-09-25  
License: CC BY 4.0 with a link back

CVE-2026-72560 · Severity: medium · CVSS 6.5 · Published 2026-08-11

Label Studio is a data labeling and annotation platform used by organizations to prepare training data. A misconfiguration in the default setup allows authenticated users to access internal network services and cloud metadata endpoints through the URL import feature, potentially exposing sensitive credentials and internal infrastructure details.

## Quick answers

**Is CVE-2026-72560 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-72560 being exploited in the wild.

**How severe is CVE-2026-72560?** Medium severity, with a CVSS score of 6.5, published 11 August 2026.

**Which products and versions are affected?** HumanSignal Label Studio through 1.24.0.dev0.

**Is there a fix for CVE-2026-72560?** No fixed version is recorded yet. Check the vendor's advisory in the references below. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2026-72560 attacked?** It is attacked over the network.

**Am I affected by CVE-2026-72560?** Check whether anything you run is named in the advisory. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## Technical details

This vulnerability is a server-side request forgery (SSRF) in the import-from-URL endpoint caused by SSRF_PROTECTION_ENABLED being set to false by default. An authenticated user can supply arbitrary URLs, including loopback addresses (127.0.0.1) and internal cloud metadata endpoints (e.g., AWS IMDSv1), which the server will fetch and return. The attack requires authentication but no additional preconditions. An attacker can reach internal services not intended for external access, enumerate network topology, and exfiltrate cloud credentials. No information on patch availability is provided in the advisory.

## Affected products

- HumanSignal Label Studio through 1.24.0.dev0

## Timeline

- 2026-08-11: disclosed

## References

- [https://github.com/HumanSignal/label-studio](https://github.com/HumanSignal/label-studio)