CVE-2024-4181
HIGH8.8EPSS 1.6%RunGptLLM class in LlamaIndex has a command injection
Published: 5/16/2024Modified: 5/19/2024
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.
Affected packages (2)
- PyPI/llama-indexfrom 0, < 0.10.13
- PyPI/llama-index-llms-rungptfrom 0, < 0.1.3
CVSS scores
| Source | Version | Severity | Vector |
|---|---|---|---|
| osv | CVSS 3.1 | HIGH8.8 | CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H |