NewApp Monitoring is liveSign up free
AgentbeamBeta
Open-source package vulnerability · PyPI

GHSA-w476-p2h3-79g9: uv has differential in tar extraction with PAX headers

low

Impact In versions 0.9.4 and earlier of uv, tar archives containing PAX headers with file size overrides were not handled properly. As a result, an attacker could contrive a source distribution (as a tar archive) that would extract differently when installed via uv versus other Python package installers.

Published Oct 21, 2025

Affected packages

uv

Ecosystem
PyPI
Affected
<= 0.9.4
Fixed in
0.9.5
Check
pip show uv

Advisory details

From the GitHub Advisory Database (CC BY 4.0).

Impact

In versions 0.9.4 and earlier of uv, tar archives containing PAX headers with file size overrides were not handled properly. As a result, an attacker could contrive a source distribution (as a tar archive) that would extract differently when installed via uv versus other Python package installers.

The underlying parsing differential here originates with astral-tokio-tar, which disclosed this vulnerability as CVE-2025-62518.

In practice, the impact of this vulnerability is low: only source distributions can be formatted as tar archives, and source distributions execute arbitrary code at build/installation time by definition. Consequently, a parser differential in tar extraction is strictly less powerful than the capabilities already exposed to an attacker who has the ability to control source distributions.

However, this particular source of malleability in source distributions is unintentional and not operating by design, and therefore we consider it a vulnerability despite its overlap in capabilities with intended behavior.

Patches

Versions 0.9.5 and newer of uv address the vulnerability above. Users should upgrade to 0.9.5 or newer.

Workarounds

Users are advised to upgrade to version 0.9.5 or newer to address this advisory.

Users should experience no breaking changes as a result of the patch above.

References

  • See CVE-2025-62518 for the corresponding advisory against astral-tokio-tar

Credited to woodruffw, zanieb.

References

How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology

More PyPI vulnerabilities

All PyPI entries