cve.bz

CVE-2026-107287

6.5 mediumEPSS 0.3%

Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.77.0 until 1.107.7 and 2.52.0, the local web_fetch_tool and the WebFetch local fallback can consume excessive CPU and memory during HTML-to-Markdown conversion of attacker-controlled HTML containing deeply nested block elements. Conversion repeatedly reprocesses accumulated text and can greatly expand intermediate output before the returned-content limit is applied, allowing a model-directed fetch to delay other work in the process. Provider-native web fetching is not affected. This issue is fixed in versions 1.107.7 and 2.52.0.

CVSS 3.1
6.5CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H
EPSS
0.3% chance of exploitation in the next 30 days; higher than 19% of CVEs (scores of 2026-10-25)
CWEs
CWE-400, CWE-407
Published
Modified
Record from
the NVD, synced

Affected

Packages (OSV)

AdvisoryEcosystemPackageAffectedFixed in
GHSA-v36g-jcw9-x7cwPyPIpydantic-ai>=1.77.0 <1.107.71.107.7
GHSA-v36g-jcw9-x7cwPyPIpydantic-ai>=2.0.0b1 <2.52.02.52.0
GHSA-v36g-jcw9-x7cwPyPIpydantic-ai-slim>=1.77.0 <1.107.71.107.7
GHSA-v36g-jcw9-x7cwPyPIpydantic-ai-slim>=2.0.0b1 <2.52.02.52.0
PYSEC-2026-4281PyPIpydantic-ai>=1.77.0 <1.107.7
>=2.0.0b1 <2.52.0
1.107.7, 2.52.0
PYSEC-2026-4290PyPIpydantic-ai-slim>=1.77.0 <1.107.7
>=2.0.0b1 <2.52.0
1.107.7, 2.52.0

References

Check your stackJSON