Vulnerability · Published Jul 8, 2026 · Assigned by VulDB
CVE-2026-15035bentoml OpenLLM Model Repository Directory Name common.py async_run_command command injection
Medium 5.3
In plain English
A medium-severity command injection in OpenLLM from bentoml. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 2.2% (higher than 83% of all known vulnerabilities).
Official 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.
Affected packages & fixed versions
| Ecosystem | Package | Fixed in | Advisory |
|---|---|---|---|
| PyPI | bentoml | no fix listed | PYSEC-2026-2089 |
Affected products (as reported)
Weakness: CWE-77, CWE-74 · 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
References
- VDB-376786 | bentoml OpenLLM Model Repository Directory Name common.py async_run_command command injection vdb-entry, technical-description
- VDB-376786 | CTI Indicators (IOB, IOC, TTP, IOA) signature, permissions-required
- CVE-2026-15035 | CVE Analysis and Report third-party-advisory
- Submit #850895 | Bentoml OpenLLM 0.6.30 Command Injection third-party-advisory
- https://github.com/bentoml/OpenLLM/issues/1229 exploit, issue-tracking
- https://github.com/bentoml/OpenLLM/pull/1235 issue-tracking, patch
- https://github.com/bentoml/OpenLLM/ product