The vulnerability of the extractFromZipFile() function in the model.go package of the Ollama system, which is used to run and manage large language models (LLMs), allows a malicious actor to influence the confidentiality and integrity of the protected information.

The vulnerability of the extractFromZipFile function in the model.go package of the Ollama system, which is used to run and manage large language models, is related to an incorrect path limitation for the restricted-access directory. Exploiting this vulnerability could allow a malicious actor to...