CVE-2024-4181
highpriority · track · 1EPSS 0.021 · 82 %ileEUUSEPSS · scored 2026-10-27
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.
cvss
Technical severity, 0–10. One row per scoring source.
base score 0–10 · higher = more severe · hover a vector to decode
ssvc
A stakeholder decision — act, attend, or track — from exploitation and impact.
weaknesses
The CWE weakness classes this flaw belongs to.
affected products (5)
Products and version ranges known to be affected.
advisories (3)
Vendor and ecosystem advisories describing this.
- GHSA-pw38-xv9x-h8chRunGptLLM class in LlamaIndex has a command injection2024-05-16ghsa · osvUS
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.
- PYSEC-2024-3242024-05-16vendor · osvUS
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.
- PYSEC-2026-1565RunGptLLM class in LlamaIndex has a command injection2026-07-07vendor · osvUS
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.
also known as (8)
Other names and identifiers for this entity.
sources
Which sources assert this record, and when.
references (7)
External citations — patches, advisories, analysis.
- https://cxsecurity.com/cveshow/CVE-2024-4181/cnnvd
- https://github.com/run-llama/llama_index/commit/d73715eaf0642705583e7897c78b9c8dd2d3a7bacnnvd
- https://huntr.com/bounties/1a204520-598a-434e-b13d-0d34f2a5ddc1cnnvd
- https://github.com/run-llama/llama_index/commit/d73715eaf0642705583e7897c78b9c8dd2d3a7baenisa_euvd
- https://huntr.com/bounties/1a204520-598a-434e-b13d-0d34f2a5ddc1enisa_euvd
- https://github.com/run-llama/llama_index/commit/d73715eaf0642705583e7897c78b9c8dd2d3a7baPatchnvd
- https://huntr.com/bounties/1a204520-598a-434e-b13d-0d34f2a5ddc1ExploitThird Party Advisorynvd
Attackers weaponise vulnerabilities like this against internet-exposed services. See what is exposed on your perimeter with a free passive scan.
Free scan