CVE-2026-107291
Description
Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 0.3.4 until 1.107.6 and 2.44.0, OpenTelemetry instrumentation configured with InstrumentationSettings(include_content=False) can still export sensitive agent content through exception.message and exception.stacktrace events, error status descriptions, and model_request_parameters containing instructions or the prompted_output_template. The exposed data is available to readers of the configured telemetry backend and can include tool feedback, provider error bodies, runtime instructions, and structured-output templates even though message attributes are redacted. This issue does not grant new access to agent data, and deployments that do not use include_content=False are not affected by the setting bypass. This issue is fixed in versions 1.107.6 and 2.44.0.
How to fix
- pydantic pydantic_ai1.107.4, 2.44.0 or later
- pydantic-ai1.107.6, 2.44.0 or later
- pydantic-ai-slim1.107.6, 2.44.0 or later
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| pydantic pydantic_ai | 0.3.4 ≤ v < 1.107.4, 2.0.0 ≤ v < 2.44.0 | 1.107.4, 2.44.0 |
| pydantic-aiPyPI | 0.3.4 ≤ v < 1.107.6, 2.0.0b1 ≤ v < 2.44.0 | 1.107.6, 2.44.0 |
| pydantic-ai-slimPyPI | 0.3.4 ≤ v < 1.107.6, 2.0.0b1 ≤ v < 2.44.0 | 1.107.6, 2.44.0 |
Check your version
Choose the product and enter the version you run to see whether it falls in the affected range.
Attack Surface Management
See the assets tied to your domains, the technologies they run and known vulnerabilities in one list.