cve/2024/CVE-2024-4181.md

### [CVE-2024-4181](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2024-4181)
![](https://img.shields.io/static/v1?label=Product&message=run-llama%2Fllama_index&color=blue)
![](https://img.shields.io/static/v1?label=Version&message=0.9.47%20&color=brightgreen)
![](https://img.shields.io/static/v1?label=Version&message=unspecified%20&color=brightgreen)
![](https://img.shields.io/static/v1?label=Vulnerability&message=CWE-94%20Improper%20Control%20of%20Generation%20of%20Code&color=brightgreen)

### Description

A command injection vulnerability exists in the RunGptLLM class of the llama_index library, version 0.9.47, used by the RunGpt framework from JinaAI to connect to Language Learning Models (LLMs). The vulnerability arises from the improper use of the eval function, allowing a malicious or compromised LLM hosting provider to execute arbitrary commands on the client's machine. This issue was fixed in version 0.10.13. The exploitation of this vulnerability could lead to a hosting provider gaining full control over client machines.

### POC

#### Reference
No PoCs from references.

#### Github
- https://github.com/XiaomingX/nice-juejin-article
- https://github.com/XiaomingX/weekly
- https://github.com/yux1azhengye/yux1azhengye