Tarfile.extract() does not fully honor the filter parameter when extracting hardlinks, allowing a crafted tar archive to result in files written with arbitrary UID/GID values. Even when callers specify filter='data', the extraction process still uses the ownership information embedded in the archive. This breach of file ownership integrity can enable an attacker to create files owned by any user or group, potentially leading to privilege escalation or manipulation of protected resources.

Python Software Foundation CPython, any environment that imports the tarfile module and invokes the extract() method on untrusted tar archives. No specific CPython release numbers are cited in the advisory, so all supported CPython versions that use the affected code path are potentially vulnerable until the official patch is released.

The CVSS score of 2 indicates a low severity assessment. EPSS data is not available, and the vulnerability is not listed in the CISA KEV catalog, suggesting it has not been actively exploited in the wild. The attack vector is likely local: an adversary needs to supply a malicious tar file to an application that performs extraction. No network-facing exposure is described, so the risk is confined to environments that process untrusted tar content. Nevertheless, because the flaw allows unauthorized file ownership changes, it can undermine isolation boundaries in multi‑tenant or privileged services.