The vulnerability of the get_cache_dir() function in the framework for working with Large Language Models (LLM) like LlamaIndex allows a attacker to execute a cache poisoning attack.

The vulnerability of the getcachedir function in the framework for working with Large Language Models LLM like LlamaIndex is related to the creation of temporary files with insecure permissions. Exploiting this vulnerability could allow an attacker to execute a cache poisoning attack...