Remote code execution via Zip Slip vulnerability in model download

Summary

Docling simplifies document processing by parsing diverse formats and providing integrations with the generative AI ecosystem. Prior to 2.91.0, the EasyOCR model download functionality extracted ZIP archives without validating member paths, enabling Zip Slip attacks. If an attacker could compromise the model download source (via supply chain attack, DNS spoofing, or MITM), they could write arbitrary files to any location writable by the process, potentially achieving remote code execution by overwriting Python files or system binaries, persistent backdoors by modifying startup scripts or SSH keys, and data corruption or system compromise. This vulnerability is fixed in 2.91.0. A flaw was found in Docling. The EasyOCR model download functionality improperly extracts ZIP archives without validating member paths, enabling Zip Slip attacks. If an attacker compromises the model download source (e.g., via a supply chain or Man-in-the-Middle (MITM) attack), they could write arbitrary files to the system. This could lead to remote code execution, allowing for persistent backdoors, data corruption, or full system compromise. This is an Important vulnerability in Docling, affecting Red Hat OpenShift AI, where the EasyOCR model download functionality is susceptible to Zip Slip attacks. An attacker who compromises the model download source, such as through a supply chain or Man-in-the-Middle (MITM) attack, could write arbitrary files to the system. This could lead to remote code execution, allowing for persistent backdoors, data corruption, or full system compromise. Red Hat severity: Important — CVSS 8.3 (CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:H/A:H). Weakness: CWE-22. No fixing RHSA erratum has published yet; monitor the Red Hat CVE page and patch when it ships.