CVE-2026-133505
Tool result data shaped like a file is rebuilt as a file download or provider file reference when message history is reloaded
Description
When message history is serialized and loaded again, a tool result that contains a JSON object in the serialized shape of one of Pydantic AI's file types is rebuilt as that file. A third party who controls data a tool returns, such as a record that an MCP server or a function tool returns from an external system, can use this to make the application download a URL on its local network, or make the model provider read an s3:// object or uploaded file with the application's credentials. Applications are affected when tool results can contain third-party data, message history containing those results is serialized and loaded again (stored history, history a UI adapter holds on the server, or Temporal, DBOS or Prefect durable execution), and the model is one Pydantic AI downloads files for, one that honors force_download, or one whose provider resolves non-HTTP file URLs or uploaded files. History that a client submits through the Vercel AI or AG-UI adapter in the default mode is not affected. The issue is fixed in pydantic-ai and pydantic-ai-slim 1.107.10 and 2.66.0.
INFO
Published Date :
Nov. 20, 2026, 8:18 a.m.
Last Modified :
Nov. 20, 2026, 8:18 a.m.
Remotely Exploit :
Yes !
Source :
7cec5f2f-3032-4006-8740-31f605ffab5f
CVSS Scores
| Score | Version | Severity | Vector | Exploitability Score | Impact Score | Source |
|---|---|---|---|---|---|---|
| CVSS 3.1 | MEDIUM | MITRE-CVE | ||||
| CVSS 3.1 | MEDIUM | 7cec5f2f-3032-4006-8740-31f605ffab5f |
Solution
- Update pydantic-ai and pydantic-ai-slim to version 1.107.10.
- Update pydantic-ai and pydantic-ai-slim to version 2.66.0.
- Ensure tool results do not contain untrusted third-party data.
- Review file handling configurations within the application.
References to Advisories, Solutions, and Tools
Here, you will find a curated list of external links that provide in-depth
information, practical solutions, and valuable tools related to
CVE-2026-133505.
| URL | Resource |
|---|---|
| https://github.com/pydantic/pydantic-ai/security/advisories/GHSA-4cj4-hgwf-4hc4 | |
| https://pydantic.dev/docs/ai/ui/overview/#trust-model-for-client-submitted-messages |
CWE - Common Weakness Enumeration
While CVE identifies
specific instances of vulnerabilities, CWE categorizes the common flaws or
weaknesses that can lead to vulnerabilities. CVE-2026-133505 is
associated with the following CWEs:
Common Attack Pattern Enumeration and Classification (CAPEC)
Common Attack Pattern Enumeration and Classification
(CAPEC)
stores attack patterns, which are descriptions of the common attributes and
approaches employed by adversaries to exploit the CVE-2026-133505
weaknesses.
We scan GitHub repositories to detect new proof-of-concept exploits. Following list is a collection of public exploits and proof-of-concepts, which have been published on GitHub (sorted by the most recently updated).
Results are limited to the first 15 repositories due to potential performance issues.
The following list is the news that have been mention
CVE-2026-133505 vulnerability anywhere in the article.
The following table lists the changes that have been made to the
CVE-2026-133505 vulnerability over time.
Vulnerability history details can be useful for understanding the evolution of a vulnerability, and for identifying the most recent changes that may impact the vulnerability's severity, exploitability, or other characteristics.
-
New CVE Received by 7cec5f2f-3032-4006-8740-31f605ffab5f
Nov. 20, 2026
Action Type Old Value New Value Added Description When message history is serialized and loaded again, a tool result that contains a JSON object in the serialized shape of one of Pydantic AI's file types is rebuilt as that file. A third party who controls data a tool returns, such as a record that an MCP server or a function tool returns from an external system, can use this to make the application download a URL on its local network, or make the model provider read an s3:// object or uploaded file with the application's credentials. Applications are affected when tool results can contain third-party data, message history containing those results is serialized and loaded again (stored history, history a UI adapter holds on the server, or Temporal, DBOS or Prefect durable execution), and the model is one Pydantic AI downloads files for, one that honors force_download, or one whose provider resolves non-HTTP file URLs or uploaded files. History that a client submits through the Vercel AI or AG-UI adapter in the default mode is not affected. The issue is fixed in pydantic-ai and pydantic-ai-slim 1.107.10 and 2.66.0. Added CVSS V3.1 AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N Added CWE CWE-918 Added CWE CWE-843 Added Affected New affected value received. <a href="https://github.com/CVEProject/cvelistV5/blob/main/cves/2026/133xxx/CVE-2026-133505.json">CVE-2026-133505</a> Added Reference https://github.com/pydantic/pydantic-ai/security/advisories/GHSA-4cj4-hgwf-4hc4 Added Reference https://pydantic.dev/docs/ai/ui/overview/#trust-model-for-client-submitted-messages