Heartex - CraftedSignal Threat Feedhttps://feed.craftedsignal.io/vendors/heartex/Trending threats, MITRE ATT&CK coverage, and detection metadata. Fed continuously.Hugoenhello@craftedsignal.iohello@craftedsignal.ioMon, 24 Aug 2026 20:06:10 +0000https://feed.craftedsignal.io/favicon-32x32.pngCraftedSignal Threat Feedhttps://feed.craftedsignal.io/3232https://feed.craftedsignal.io/favicon.svgIDOR Vulnerability in Label Studio Annotation APIhttps://feed.craftedsignal.io/briefs/2026-08-label-studio-idor/Mon, 24 Aug 2026 20:06:10 +0000hello@craftedsignal.iohttps://feed.craftedsignal.io/briefs/2026-08-label-studio-idor/Label Studio contains an insecure direct object reference (IDOR) vulnerability, CVE-2026-76073, allowing authenticated users to read, modify, or delete annotations across organizational boundaries by enumerating sequential identifiers.Label Studio, a popular open-source data labeling tool maintained by Heartex, contains a critical insecure direct object reference (IDOR) vulnerability identified as CVE-2026-76073. The flaw exists within the 'AnnotationAPI' and 'AnnotationConvertAPI' endpoints defined in 'label_studio/tasks/api.py'. The application fails to properly scope querysets to the requesting user's organization. While other endpoints, such as the task-specific API, correctly constrain data access by checking 'project__organization', the annotation endpoints default to 'Annotation.objects.all()'. Because these endpoints only verify basic authentication rather than enforcing object-level authorization, any logged-in user can access any annotation in the system. Furthermore, since annotation identifiers are assigned as sequential integers, an attacker can trivially enumerate these IDs to exfiltrate or manipulate sensitive annotation data belonging to other organizations hosted on the same instance. This vulnerability poses a significant risk to the confidentiality and integrity of proprietary datasets within multi-tenant Label Studio environments.

Impact

The vulnerability allows any authenticated user to gain unauthorized access to annotation data across organizational silos. Impact includes the potential for full data exfiltration of labels, the modification of labeling tasks which could poison machine learning pipelines, and the deletion of project work. This is particularly critical in multi-tenant environments where organizations expect strict data isolation.

Recommendation

  1. Upgrade Label Studio to a patched version once provided by Heartex to remediate CVE-2026-76073.
  2. Implement an application-layer firewall or web application firewall (WAF) to monitor for anomalous spikes in API requests targeting the annotation endpoints (e.g., sequentially incrementing IDs).
  3. Audit webserver access logs for high-volume HTTP GET, PATCH, and DELETE requests to '/api/annotations/' from low-privileged user accounts.
  4. Ensure that all organizational members are vetted and that the instance is restricted to trusted users until the patch is applied.
]]>highadvisoryidorapi-securitydata-exfiltration