Intel

AIKIDO-2025-10414

pydantic-ai-slim is vulnerable to Improper Removal of Sensitive Information Before Storage or Transfer

Improper Removal of Sensitive Information Before Storage or Transfer Pre-CVE
Found by Aikido Intel before public disclosure or CVE publication.

15

Low Risk

This Affects:

PYTHONpydantic-ai-slim
0.0.1 - 0.3.3
Fixed in 0.3.4

TL;DR

Affected versions of pydantic-ai-slim may inadvertently expose sensitive information to observability platforms. This includes user data, proprietary prompts, tool call arguments, and model responses captured in OpenTelemetry events. To address this, patched versions are able prevent sensitive content from being included in telemetry data, enabling secure monitoring of agent behavior and performance without compromising user privacy or proprietary information.

Who does this affect?

You are affected if you are using a vulnerable version of the package.

Background info

pydantic-ai-slim is vulnerable to Improper Removal of Sensitive Information Before Storage or Transfer in versions 0.0.1 - 0.3.3.

How to fix this

Upgrade pydantic-ai-slim to the patch version and set include_content=False or make sure no OpenTelemetry events are sent in your production environment.

Background Info