CVE-2026-65975
Pydantic AI AG-UI Adapter: A dangling client-submitted tool call can execute when a trailing message is dropped during sanitize_messages
Texto original en inglés · Ver en español
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N
Fuente: CNA
Ciclo de vida de la vulnerabilidad
| Fecha | Fase | Evento | Detalle |
|---|---|---|---|
| Remediación | Parche 2.5.0 | pydantic-ai · -25 d desde divulgación | |
| Remediación | Parche 1.107.1 | pydantic-ai · -18 d desde divulgación | |
| Divulgación | Publicada | CVE-2026-65975 divulgada · CVSS 6.5 (CNA) | |
| Divulgación | Evaluación SSVC | CISA-ADP · explotación ninguna | |
| Divulgación | Modificada | Última actualización del registro | |
| Monitoreo | Detectada | Ingresó al índice | |
| Monitoreo | Fuente: OSV | Registro visto por primera vez en OSV | |
| Monitoreo | Verificada | Última verificación contra las fuentes |
Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. In versions 1.88.0 up to but not including 1.107.1 and 2.0.0b1 up to but not including 2.5.0, the UI adapters (AG-UI via Agent.to_ag_ui()/AGUIAdapter, and Vercel AI via VercelAIAdapter) use sanitize_messages to strip unresolved ("dangling") client-submitted tool calls from untrusted message history before it reaches the agent, a defense-in-depth default that prevents the agent from executing tool calls the model never emitted. However, the strip anchored to a message index computed before sanitization ran, so when a trailing client message sanitized to empty and was dropped (for example a client system message under the default manage_system_prompt='server'), a preceding assistant response carrying an unresolved tool call became the new tail and was dispatched without inspection. As a result, a remote client could cause a registered, non-approval server tool to run with client-supplied arguments rather than arguments the model produced. The impact is bounded by what the affected tools do and is most significant for applications that gate tool execution in a model-request hook (before_model_request / after_model_request), since a forged call skips the model turn and bypasses that guardrail; approval-gated tools (requires_approval=True) are not auto-executed by this path. This issue has been fixed in versions 1.107.1 and 2.5.0.
Desglose de métricas
Rangos afectados
Estado del parche por paquete
| Paquete / producto | Estado | Corregida en | Desde divulgación | Última versión |
|---|---|---|---|---|
pypipydantic-ai |
Corregida | 1.107.1, 2.5.0 | -18 d · -25 d | 2.68.0 no afectada |
pypipydantic-ai-slim |
Corregida | 1.107.1, 2.5.0 | -18 d · -25 d | 2.71.0 no afectada |
1.107.12.5.0Arsenal público
Sin exploits, módulos ni PoC públicos conocidos a la fecha de verificación.
https://github.com/pydantic/pydantic-ai/security/advisories/GHSA-jpr8-2v3g-wgf9Las referencias externas se muestran para copiar: cve.llama.ec no lo redirige a otros sitios.
