CVE-2025-64512 is a deserialization vulnerability classified under CWE-502 affecting pdfminer.six, a widely used Python library for extracting information from PDF documents. The vulnerability exists in the CMapDB._load_data() function, which deserializes data using Python's pickle.loads() method. Normally, pdfminer.six loads trusted pickle files from its internal cmap/ directory. However, a malicious PDF can specify an alternative directory and filename ending with .pickle.gz, causing pdfminer.six to load and decompress a malicious zipped pickle file. Since pickle deserialization can execute arbitrary code, this leads to remote code execution when the PDF is processed. The vulnerability requires no privileges but does require user interaction to open or process the malicious PDF. The scope is broad as any system using vulnerable pdfminer.six versions (< 20251107) to parse PDFs is at risk. The CVSS 3.1 score of 8.6 reflects high impact on confidentiality, integrity, and availability, with low attack complexity but requiring user interaction. No known exploits are currently reported in the wild, but the vulnerability is critical due to the nature of pickle deserialization and the common use of pdfminer.six in automated PDF workflows.

For European organizations, this vulnerability poses a significant risk especially to sectors relying heavily on automated PDF processing such as finance, legal, government, and software development. Exploitation can lead to arbitrary code execution on systems processing malicious PDFs, potentially resulting in data breaches, system compromise, lateral movement within networks, and disruption of critical services. Confidential information extracted or stored by these systems could be exposed or altered, impacting compliance with GDPR and other data protection regulations. The availability of affected systems could also be compromised, affecting business continuity. Since pdfminer.six is often integrated into larger applications or pipelines, the attack surface extends beyond standalone usage, increasing the risk of supply chain or third-party software compromise. The requirement for user interaction (opening or processing a malicious PDF) means phishing or social engineering could be used to deliver the exploit.

1. Upgrade all instances of pdfminer.six to version 20251107 or later, which contains the fix preventing untrusted pickle deserialization. 2. Implement strict input validation and restrict PDF processing to trusted sources only, minimizing exposure to malicious PDFs. 3. Employ sandboxing or containerization for PDF processing workflows to isolate potential code execution from critical systems. 4. Monitor and log PDF processing activities to detect anomalous behavior indicative of exploitation attempts. 5. Educate users about the risks of opening PDFs from untrusted or unknown sources to reduce the likelihood of social engineering attacks. 6. Review and audit any custom code or third-party tools that integrate pdfminer.six to ensure they do not override safe defaults or introduce insecure configurations. 7. Consider disabling or replacing pickle-based deserialization in custom workflows with safer alternatives if feasible.