---
id: CVE-2026-107289
title: >-
  Pydantic AI is a Python agent framework for building applications and
  workflows with Generative AI
summary: >-
  Pydantic AI is a Python agent framework for building applications and
  workflows with Generative AI. From 1.56.0 until 1.107.6 and 2.44.0,
  applications that opt attacker-influenced URLs into local network access
  through FileUrl with force…
severity: medium
cvss: 6.8
cvssVector: 'CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:N/A:N'
cwe:
  - CWE-918
  - CWE-1289
vendor: pydantic
product: pydantic_ai
affected:
  - 'pydantic_ai >= 1.56.0, < 1.107.6'
  - 'pydantic_ai >= 2.0.0, < 2.44.0'
patched:
  - pydantic_ai 2.44.0
published: '2026-10-08'
updated: '2026-10-13'
sourceUpdated: '2026-10-13T14:34:48.123'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-107289'
references:
  - url: >-
      https://github.com/pydantic/pydantic-ai/commit/02157e1b87bd45d3f2e111ce07afdf89f9fb0e5b
    label: security-advisories@github.com
  - url: >-
      https://github.com/pydantic/pydantic-ai/commit/4da70591460f51a8c4f128eaeef70a33340dbd55
    label: security-advisories@github.com
  - url: 'https://github.com/pydantic/pydantic-ai/pull/8401'
    label: security-advisories@github.com
  - url: 'https://github.com/pydantic/pydantic-ai/pull/8402'
    label: security-advisories@github.com
  - url: 'https://github.com/pydantic/pydantic-ai/releases/tag/v1.107.6'
    label: security-advisories@github.com
  - url: 'https://github.com/pydantic/pydantic-ai/releases/tag/v2.44.0'
    label: security-advisories@github.com
  - url: >-
      https://github.com/pydantic/pydantic-ai/security/advisories/GHSA-vmxc-h2x2-jmf3
    label: security-advisories@github.com
  - url: 'https://github.com/advisories/GHSA-vmxc-h2x2-jmf3'
  - url: 'https://github.com/pydantic/pydantic-ai'
  - url: >-
      https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-107289.json
  - url: 'https://access.redhat.com/security/cve/CVE-2026-107289'
  - url: 'https://bugzilla.redhat.com/show_bug.cgi?id=2548032'
  - url: 'https://www.cve.org/CVERecord?id=CVE-2026-107289'
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-107289'
  - url: 'https://pypi.org/project/pydantic-ai'
  - url: 'https://pypi.org/project/pydantic-ai-slim'
tags:
  - nvd
  - ghsa
  - pip
  - cve.org
  - osv
  - csaf
  - vex
  - red-hat
  - score-dispute
epss: 0.0039
epssPercentile: 0.32952
aliases:
  - GHSA-vmxc-h2x2-jmf3
  - PYSEC-2026-4282
  - PYSEC-2026-4291
ecosystem: pip
ssvc:
  exploitation: none
  automatable: 'no'
  technicalImpact: partial
  timestamp: '2026-10-08T16:07:09.527203Z'
scores:
  nvd: 6.8
  vendor: 3.7
  osv: 6.8
ingestedAt: '2026-10-08T16:52:14.772Z'
---

## Overview

Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.56.0 until 1.107.6 and 2.44.0, applications that opt attacker-influenced URLs into local network access through FileUrl with force_download='allow-local' or web_fetch_tool with allow_local_urls=True can bypass the cloud-metadata blocklist by appending an IPv6 zone identifier to an IPv6 metadata address. IPv6Address equality and hashing include the zone identifier, so the blocklist comparison fails even though the network stack ignores the zone on a non-link-local destination and reaches the metadata service, potentially exposing cloud IAM credentials. The opt-in settings are disabled by default, and the issue requires an IPv6-enabled environment. This issue is fixed in versions 1.107.6 and 2.44.0.

## Affected

- `pydantic_ai >= 1.56.0, < 1.107.6`
- `pydantic_ai >= 2.0.0, < 2.44.0`

## Remediation

Upgrade past the affected range:

- `pydantic_ai 2.44.0`

## Package advisory (CVE-2026-107289)

Affected packages:

- `pydantic-ai >= 1.56.0, < 1.107.6`
- `pydantic-ai >= 2.0.0b1, < 2.44.0`
- `pydantic-ai-slim >= 1.56.0, < 1.107.6`
- `pydantic-ai-slim >= 2.0.0b1, < 2.44.0`

Patched in:

- `pydantic-ai 1.107.6`
- `pydantic-ai 2.44.0`
- `pydantic-ai-slim 1.107.6`
- `pydantic-ai-slim 2.44.0`

Source: https://github.com/advisories/GHSA-vmxc-h2x2-jmf3

## Vendor advisories

- **Red Hat VEX** · Low · affected: Lightspeed Core, Red Hat Enterprise Linux AI (RHEL AI) 3 · no fix planned: Lightspeed Core, Red Hat Enterprise Linux AI (RHEL AI) 3 · updated 2026-10-15 · [vex](https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-107289.json)