CVE-2025-62373
Published: 23 April 2026
Description
Pipecat is an open-source Python framework for building real-time voice and multimodal conversational agents. Versions 0.0.41 through 0.0.93 have a vulnerability in `LivekitFrameSerializer` – an optional, non-default, undocumented frame serializer class (now deprecated) intended for LiveKit integration. The class's `deserialize()`…
more
method uses Python's `pickle.loads()` on data received from WebSocket clients without any validation or sanitization. This means that a malicious WebSocket client can send a crafted pickle payload to execute arbitrary code on the Pipecat server. The vulnerable code resides in `src/pipecat/serializers/livekit.py` (around line 73), where untrusted WebSocket message data is passed directly into `pickle.loads()` for deserialization. If a Pipecat server is configured to use LivekitFrameSerializer and is listening on an external interface (e.g. 0.0.0.0), an attacker on the network (or the internet, if the service is exposed) could achieve remote code execution (RCE) on the server by sending a malicious pickle payload. Version 0.0.94 contains a fix. Users of Pipecat should avoid or replace unsafe deserialization and improve network security configuration. The best mitigation is to stop using the vulnerable LivekitFrameSerializer altogether. Those who require LiveKit functionality should upgrade to the latest Pipecat version and switch to the recommended `LiveKitTransport` or another secure method provided by the framework. Additionally, always follow secure coding practices: never trust client-supplied data, and avoid Python pickle (or similar unsafe deserialization) in network-facing components.
Mitigating Controls (NIST 800-53 r5)AI
Requires timely identification, reporting, and correction of the deserialization flaw by upgrading to Pipecat version 0.0.94, directly eliminating the vulnerability.
Prohibits or restricts the use of the optional, non-default, deprecated LivekitFrameSerializer, preventing exposure to the unsafe deserialization in network-facing configurations.
Mandates validation of untrusted WebSocket inputs to block crafted pickle payloads that enable arbitrary code execution on the Pipecat server.
Security SummaryAI
CVE-2025-62373 is a critical deserialization vulnerability (CWE-502) affecting the Pipecat open-source Python framework for building real-time voice and multimodal conversational agents. Versions 0.0.41 through 0.0.93 contain the issue in the optional, non-default, undocumented LivekitFrameSerializer class, intended for LiveKit integration and now deprecated. Located in src/pipecat/serializers/livekit.py around line 73, the class's deserialize() method directly invokes Python's pickle.loads() on untrusted data received from WebSocket clients without validation or sanitization, enabling arbitrary code execution.
The vulnerability can be exploited by any unauthenticated attacker (AV:N/AC:L/PR:N) with network access to a Pipecat server configured to use LivekitFrameSerializer and listening on an external interface such as 0.0.0.0. A malicious WebSocket client can send a crafted pickle payload, leading to remote code execution (RCE) on the server. The CVSS v3.1 base score of 9.8 reflects high confidentiality, integrity, and availability impacts (C:H/I:H/A:H) with no user interaction required.
The GitHub security advisory (GHSA-c2jg-5cp7-6wc7) recommends upgrading to Pipecat version 0.0.94, which contains a fix. Users should immediately stop using the vulnerable LivekitFrameSerializer; those needing LiveKit functionality must switch to the recommended LiveKitTransport or another secure method in the framework. Additional mitigations include avoiding Python pickle in network-facing components, never trusting client-supplied data, and hardening network security configurations to limit exposure.
Details
- CWE(s)
Affected Products
MITRE ATT&CK Enterprise TechniquesAI
Why these techniques?
Unauthenticated RCE via unsafe deserialization (pickle.loads) in public-facing WebSocket service directly enables T1190 (Exploit Public-Facing Application) and facilitates T1059.006 (Python code execution).