CVE-2025-27772
UpTrain is an open-source platform to evaluate and improve generative AI applications. In version 0.7.1 and prior, the `/new_run` endpoint is vulnerable to remote code execution via the `checks` and `metadata` parameters. Any user that has access to UpTrain and a valid authentication method may be able to execute arbitrary code in the context of the host running UpTrain, which in most cases will be the docker container as suggested by the documentation. As of time of publication, no known patch is available.
Scoring
- Severity
- HIGH
- CVSS base score
- 7.4
- CVSS vector
- CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:P
- EPSS probability
- 0.61%
- CWE
- CWE-74
- Published
- 2026-08-17
- Last modified
- 2026-08-17
Affected products
- uptrain-ai uptrain
Weakness type
Related vulnerabilities
- CVE-2026-87572 — Injection in DevTools in Google Chrome prior to 153.0.8010.36 allowed a remote attacker who had compromised the renderer
- CVE-2026-82971 — QVidium Opera11 CGI Script net_tr.cgi command injection
- CVE-2026-20130 — Cisco Identity Services Engine Hardening Release - Improper Neutralization Vulnerabilities
- CVE-2026-12351 — IBM MQ is vulnerable to unauthenticated remote code execution via JNDI injection
- CVE-2026-79697 — Advantech WISE-6610-NB Basic Station Certificate-Deletion basicstation_apply command injection
- CVE-2026-79698 — Advantech WISE-6610-NB Node-RED nodered_lib_apply command injection
- CVE-2026-79234 — Injection in CSS in Google Chrome prior to 152.0.7977.65 allowed a remote attacker to potentially obtain sensitive infor
- CVE-2026-83772 — Cobham SATCOM VSAT7090 Maritime Satellite Router JSON Parsing mail-report.sh c_set_reports_decode command injection