https://cve.radiocsirt.org Contains only the most 10 recent entries. http://www.rssboard.org/rss-specification python-feedgen en Wed, 28 Oct 2026 07:23:09 +0000 https://cve.radiocsirt.org/vuln/cleanstart-2026-bf25173 <p><strong>Affected:</strong> CleanStart: py3-pydantic-ai</p> <p>Security vulnerability affects the py3-pydantic-ai package. Pydantic AI is a Python agent framework for building applications and workflows with Generative AI.</p> <p><strong>Affected:</strong> CleanStart: py3-pydantic-ai</p> <p>Security vulnerability affects the py3-pydantic-ai package. Pydantic AI is a Python agent framework for building applications and workflows with Generative AI.</p> https://cve.radiocsirt.org/vuln/cleanstart-2026-bf25173 https://cve.radiocsirt.org/vuln/euvd-2026-384690 EUVD-2026-384690 https://cve.radiocsirt.org/vuln/euvd-2026-384690 https://cve.radiocsirt.org/vuln/fkie_cve-2026-107288 <p>Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.77.0 until 1.107.6 and 2.44.0, the local web_fetch_tool and the WebFetch local fallback compare blocked_domains entries with a URL hostname before both values are normalized to the form used by getaddrinfo. An attacker-influenced model can use an equivalent IDNA spelling, non-ASCII label separator, case variation, or trailing root label that resolves to a blocked host but does not match the configured string, causing the application to fetch that host with its own privileges. allowed_domains fails closed for unmatched spellings, and private-IP and cloud-metadata protections remain effective. This issue is fixed in versions 1.107.6 and 2.44.0.</p> <p>Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.77.0 until 1.107.6 and 2.44.0, the local web_fetch_tool and the WebFetch local fallback compare blocked_domains entries with a URL hostname before both values are normalized to the form used by getaddrinfo. An attacker-influenced model can use an equivalent IDNA spelling, non-ASCII label separator, case variation, or trailing root label that resolves to a blocked host but does not match the configured string, causing the application to fetch that host with its own privileges. allowed_domains fails closed for unmatched spellings, and private-IP and cloud-metadata protections remain effective. This issue is fixed in versions 1.107.6 and 2.44.0.</p> https://cve.radiocsirt.org/vuln/fkie_cve-2026-107288