---
id: CVE-2024-4181
aliases:
  - GHSA-pw38-xv9x-h8ch
  - PYSEC-2026-1565
title: RunGptLLM class in LlamaIndex has a command injection
summary: RunGptLLM class in LlamaIndex has a command injection
severity: high
cvss: 8.8
cvssVector: 'CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H'
vendor: llama-index
product: llama-index
ecosystem: pip
affected:
  - llama-index < 0.10.13
  - llama-index-llms-rungpt < 0.1.3
patched:
  - llama-index 0.10.13
  - llama-index-llms-rungpt 0.1.3
published: '2024-05-16'
updated: '2026-07-07'
source: OSV
sourceUrl: 'https://osv.dev/vulnerability/GHSA-pw38-xv9x-h8ch'
references:
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2024-4181'
  - url: >-
      https://github.com/run-llama/llama_index/commit/d73715eaf0642705583e7897c78b9c8dd2d3a7ba
  - url: 'https://github.com/run-llama/llama_index'
  - url: 'https://huntr.com/bounties/1a204520-598a-434e-b13d-0d34f2a5ddc1'
tags:
  - osv
  - pip
epss: 0.02135
epssPercentile: 0.81225
ingestedAt: '2026-07-08T18:25:52.018Z'
---

## Overview

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

- `llama-index < 0.10.13`
- `llama-index-llms-rungpt < 0.1.3`

## Remediation

Upgrade to a patched release:

- `llama-index 0.10.13`
- `llama-index-llms-rungpt 0.1.3`