HighVulnerability
CVE-2023-5245: MLeap path traversal in FileUtil.extract() via TensorFlow saved_model loading
- Source
- NVD(opens in a new tab)
- Identifier
- CVE-2023-5245
- Published
- Record updated
- Affected
- ml.combust.mleap:mleap-runtime_2.12 < 0.23.1
- Fixed in
- 0.23.1
- Known exploitation
- Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
- EPSS
- 1.2%
Summary
CVE-2023-5245 affects FileUtil.extract() in MLeap, which enumerates all zip file entries and extracts each one without checking whether the file paths fall outside the intended directory. When TensorflowModel is created from a saved_model format export, apply() calls this vulnerable implementation. The source states that arbitrary file creation can directly lead to code execution.
Mitigation
The source links a patch in GitHub pull request combust/mleap#866 but does not describe the fix or name a fixed version. The source does not state a configuration change or workaround.
Related items
- MediumCVE-2026-121936: easyVoice server-side request forgery through openaiBaseUrl in /api/v1/ttsSimilar attack · NVD/CVE Database
- MediumCVE-2026-118628: MLflow get-artifacts server-side request forgery information disclosureSimilar attack · NVD/CVE Database
- HighCVE-2026-118625: MLflow Tracking Server privilege escalation via get_endpointsSimilar attack · NVD/CVE Database
- InfoWhen trust becomes the payload: Inside the threats, etc quarterly reportSimilar attack · Elastic Security Labs
- MediumCritical Langflow flaws enable authentication bypass and RCESimilar attack · CSO Online