Skip to main content

CVE-2026-133505pydantic pydantic-ai type confusion

Low priorityMedium5.9EPSS not scored yet

CVE-2026-133505 is a type confusion in pydantic pydantic-ai, rated medium (5.9 on CVSS 3.1).

Why this priority: Medium severity, with no exploitation signal.

Published · Updated · NVD status: Received

What it is

Memory holding one type of object is treated as another, often a route to code execution in browsers and script engines. The server can be made to fetch an address of the attacker's choosing, including internal services and cloud metadata endpoints.

How it is reached

An attacker can exploit it over the network, without an account, and nobody has to click anything. Success also depends on conditions the attacker does not fully control.

What an attacker gains

A successful attack can read the data it protects. In practice, this kind of flaw lets an attacker crash the process, read memory, or in the worst case run code.

Is it being exploited?

There is no public report of exploitation yet.

What to do

  1. No fixed version is listed yet. Find every place pydantic pydantic-ai runs, watch the references for a fix, and plan a replacement if it is no longer supported.
  2. While there is no fix, keep it off the internet or behind a VPN, allowlist or authenticating proxy.

Description from NVD

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.

What this means for your architecture.

A CVE names a product. Whether it matters depends on where that product sits in your systems: what reaches it, which trust boundary it guards and what it can reach once compromised. Veriom maps your components, their relationships and your trust boundaries from your code, cloud and delivery evidence, so you can see where a component like this sits. The radar itself does not look at your systems.

Components it touches

  • Native code: operating systems, browsers and firmware. The flaw sits wherever native code reads untrusted data.
  • Network appliances and media or file parsers. The flaw sits wherever native code reads untrusted data.
  • Services that fetch URLs or call other hosts for a caller. The flaw sits between the application and the internal network behind it.

Trust boundaries in play

  • Internet-facing entry points. No account is needed. Any instance reachable from outside your perimeter, through a load balancer, gateway, VPN or exposed admin port, is in scope.
  • Wherever native code reads untrusted data. An attacker who crosses it can crash the process, read memory, or in the worst case run code.
  • Between the application and the internal network behind it. An attacker who crosses it can make the server send requests into internal networks or cloud metadata.

Questions to answer about your own systems

  • Where does pydantic pydantic-ai run in your systems, and at which version?
  • Which of those instances can be reached from the internet or partner networks?
  • What can it reach once compromised: which data stores, secrets and services?

14-day free trial

Understand how your systems connect.

See your services, dependencies and trust boundaries in one place, with evidence from your connected environment.

Sources

  • CVE records and CVSS scores from the NVD CVE API. This product uses the NVD API but is not endorsed or certified by the NVD.
  • Exploitation status from CISA’s Known Exploited Vulnerabilities catalog.
  • Exploit likelihood from FIRST’s Exploit Prediction Scoring System (EPSS), refreshed daily.
  • Explanations are written by Veriom from fixed templates over each record’s data, not by a person reviewing each CVE. Descriptions are quoted from the records as published.