Microsoft Threat Intelligence has identified that hackers have embedded malicious code into Mistral AI software packages distributed through the Python Package Index (PyPI). The malware, which targets Linux systems, downloads a file named transformers.pyz from a remote server and executes it in the background. This malicious code primarily acts as a credential stealer, gathering developers' login credentials and access tokens. Mistral has acknowledged that a developer device was compromised due to the TanStack security incident. However, the company assures that there is no evidence of any breach in Mistral's infrastructure itself.