CVE-2025-27772
HighCVSS 7.4Exploitation Probability (EPSS)
Elevated risk56th percentile - higher than 56% of all known CVEs
Summary
In UpTrain version 0.7.1 and prior, the /new_run endpoint is vulnerable to remote code execution via the checks and metadata parameters. Any user with 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.
Risk Assessment
Remote code execution could lead to full compromise of the container or host, potentially resulting in data confidentiality and integrity breaches.
Recommendation
No known patch is available; restrict access to UpTrain and monitor activity on the /new_run endpoint.
Other vulnerabilities in UpTrain
See all- CVE-2025-27771High
In UpTrain version 0.7.1 and prior, the /add_prompts endpoint is vulnerable to remote code execution via the checks and metadata parameters. Any user with 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.
- CVE-2025-27770High
UpTrain version 0.7.1 and earlier contains a remote code execution vulnerability in the `/create_project` endpoint via the `checks` and `metadata` parameters. Any user with access to UpTrain and valid authentication can execute arbitrary code in the context of the host running UpTrain, usually a Docker container. No patch is available as of publication.
- CVE-2025-27621High
UpTrain version 0.7.1 and earlier creates a default user with a static username, which is also used as the default API key. The backend has an open CORS policy, allowing any website to make authenticated cross-origin requests using the default API key in the `uptrain-access-token` header. This may allow arbitrary websites to perform privileged operations on the UpTrain instance.
Original NVD description (English source)
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.
Vulnerability data from NVD (NIST) · CISA KEV · EPSS