Introducing The Sweep: OX’s New Live Security Research Series
Register Now
Breaking: The 2026 Vulnpocalypse Hits PHP: Four Critical RCEs, but How Dangerous Are They?
Read the Report
Are you exposed? Connect a repo in minutes.
Scan your repos free

CVE-2026-107287

pydantic-ai: pydantic-ai: Denial of Service via resource exhaustion during HTML conversion

Overview

Published
8 Oct 2026
Modified
14 Oct 2026
Medium CVE
2 Registry packages
0 OS packages
0 Linux distributions
Yes Fix available

Description

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.

Source: OX vulnerability database

Severity

6.5 · MEDIUM

CVSS 3.1 base score, scored by NVD.

Attack vector
Network
Attack complexity
Low
Privileges required
None
User interaction
Required
Scope
Unchanged
Confidentiality
None
Integrity
None
Availability
High

CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H

Weakness: CWE-400, CWE-407

Affected packages · 2

Packages affected by CVE-2026-107287
EcosystemPackageVulnerableFixed in
PyPI pydantic-ai >=1.77.0, <1.107.7>=2.0.0b1, <2.52.0 1.107.72.52.0
PyPI pydantic-ai-slim >=1.77.0, <1.107.7>=2.0.0b1, <2.52.0 1.107.72.52.0

Open a package to see all its vulnerabilities.

Operating systems

No Linux distribution reports this ID.

References

OX Radar and Response Center

Find threats in your stack.

OX Radar can check high-priority threats, like critical package CVEs, against your context.

Each incident is marked Detected or Not Detected with a remediation plan and proof all tracked in the Response Center.