Description
A vulnerability was found in bentoml OpenLLM 0.6.30. This affects the function async_run_command of the file src/openllm/common.py of the component Model Repository Directory Name Handler. Performing a manipulation of the argument cmd results in command injection. Attacking locally is a requirement. The exploit has been made public and could be used. The project was informed of the problem early through an issue report but has not responded yet.
Published: 2026-07-08
Score: 4.8 Medium
EPSS: 1.5% Low
KEV: No
Impact: n/a
Action: n/a
AI Analysis

Impact

This vulnerability exists in the async_run_command function of the Model Repository Directory Name Handler in bentoml OpenLLM. By manipulating the cmd argument, a local attacker can inject and execute arbitrary shell commands, effectively gaining control over the system running the service. The weakness corresponds to CWE-74 and CWE-77, indicating improper command validation and potential injection.

Affected Systems

The affected product is bentoml OpenLLM version 0.6.30. No other versions or vendor products are listed as affected.

Risk and Exploitability

The CVSS score of 4.8 indicates moderate severity. The EPSS score of 2% indicates a low probability of exploitation, though a public exploit exists. The vulnerability requires local access to the target and is not listed in the CISA KEV catalog. Because the exploit is publicly disclosed, a local attacker who can reach the service could potentially execute arbitrary commands on the underlying host.

Generated by OpenCVE AI on July 26, 2026 at 17:51 UTC.

Remediation

No vendor fix or workaround currently provided.

OpenCVE Recommended Actions

  • Check for an updated OpenLLM release that contains a fix for this command-injection issue and upgrade when available.
  • If no patch is immediately available, harden the environment by restricting local access to the OpenLLM service (e.g., firewall rules, container isolation, or running the service under a low-privilege user).
  • Implement input validation to sanitize the cmd argument before it is passed to the shell, ensuring that no shell metacharacters or unintended arguments can be injected.

Generated by OpenCVE AI on July 26, 2026 at 17:51 UTC.

Tracking

Sign in to view the affected projects.

Advisories

No advisories yet.

History

Wed, 08 Jul 2026 15:30:00 +0000

Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'poc', 'Technical Impact': 'partial'}, 'version': '2.0.3'}


Wed, 08 Jul 2026 14:15:00 +0000

Type Values Removed Values Added
Description A vulnerability was found in bentoml OpenLLM 0.6.30. This affects the function async_run_command of the file src/openllm/common.py of the component Model Repository Directory Name Handler. Performing a manipulation of the argument cmd results in command injection. Attacking locally is a requirement. The exploit has been made public and could be used. The project was informed of the problem early through an issue report but has not responded yet.
Title bentoml OpenLLM Model Repository Directory Name common.py async_run_command command injection
First Time appeared Bentoml
Bentoml openllm
Weaknesses CWE-74
CWE-77
CPEs cpe:2.3:a:bentoml:openllm:*:*:*:*:*:*:*:*
Vendors & Products Bentoml
Bentoml openllm
References
Metrics cvssV2_0

{'score': 4.3, 'vector': 'AV:L/AC:L/Au:S/C:P/I:P/A:P/E:POC/RL:ND/RC:UR'}

cvssV3_0

{'score': 5.3, 'vector': 'CVSS:3.0/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L/E:P/RL:X/RC:R'}

cvssV3_1

{'score': 5.3, 'vector': 'CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L/E:P/RL:X/RC:R'}

cvssV4_0

{'score': 4.8, 'vector': 'CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P'}


cve-icon MITRE

Status: PUBLISHED

Assigner: VulDB

Published:

Updated: 2026-07-08T14:20:35.536Z

Reserved: 2026-07-08T07:41:07.309Z

Link: CVE-2026-15035

cve-icon Vulnrichment

Updated: 2026-07-08T14:20:32.251Z

cve-icon NVD

No data.

cve-icon Redhat

No data.

cve-icon OpenCVE Enrichment

Updated: 2026-07-26T18:00:05Z

Weaknesses
  • CWE-74

    Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')

  • CWE-77

    Improper Neutralization of Special Elements used in a Command ('Command Injection')