Which versions of UpTrain are affected by CVE-2025-27770?

UpTrain v0.7.1 and earlier contain a remote code execution vulnerability in the dashboard backend that allows authenticated users to execute arbitrary code on host systems.

How to fix or mitigate CVE-2025-27770?

Within 24 hours, catalog all UpTrain v0.7.1 and prior deployments in your environment and immediately restrict network access to dashboard endpoints through firewall rules or network segmentation; if the vulnerable `/create_project` API endpoint is not operationally required, disable it via configuration. Within 7 days, deploy comprehensive monitoring and alerting on all `/create_project` endpoint access attempts and authentication events, restrict dashboard API access to named service accounts with minimal required permissions, and identify the target patched UpTrain version for your upgrade path once released. Within 30 days, upgrade to a patched UpTrain version from the vendor, decommission deprecated UpTrain instances, and conduct security review of API design practices across your application portfolio to identify similar unsanitized parameter vulnerabilities.

UpTrain CVE-2025-27770

Q: What is the CVSS score of CVE-2025-27770?

CVE-2025-27770 has a CVSS 4.0 base score of 7.4 (High). 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/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X. EPSS exploitation probability: 0.8%.

EUVD-2025-210734 HIGH

Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection') (CWE-74)

2026-08-17 security-advisories@github.com

Docker RCE

7.4

CVSS 4.0 · Vendor: github

Severity by source

Vendor (github) PRIMARY

7.4 HIGH

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/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X

vuln.today AI

8.8 HIGH

Network-exploitable with low complexity by any authenticated low-privilege user (PR:L); full C/I/A reflects arbitrary code execution; scope unchanged given container-contained impact.

3.1 AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

4.0 AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

Primary rating from Vendor (github).

CVSS VectorVendor: github

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/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X

Attack Vector

Network

Attack Complexity

Low

Privileges Required

Low

User Interaction

None

Scope

Lifecycle Timeline

Analysis Generated

Aug 17, 2026 - 16:31 vuln.today

CVE Published

Aug 17, 2026 - 16:16 cve.org

HIGH 7.4

DescriptionCVE.org

UpTrain is an open-source platform to evaluate and improve generative AI applications. In version 0.7.1 and prior, the /create_project 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.

Articles & Coverage 2

News 4 New Docker Vulnerabilities - 4 High Severity, 1 With POC Aug 18, 2026 News 3 New Docker Vulnerabilities - 3 High Severity Patched Aug 17, 2026

AnalysisAI

Remote code execution in UpTrain v0.7.1 and prior is achievable by any authenticated user through the /create_project API endpoint, where the checks and metadata parameters are processed without adequate sanitization, allowing arbitrary code to run in the context of the host process. Discovered and reported by GitHub Security Lab (GHSL-2024-198/GHSL-2024-199), the flaw targets the dashboard backend (app.py:L691) and is most commonly exploited within Docker-based deployments as recommended by official documentation. No vendor patch exists as of publication, and a proof-of-concept is confirmed by the CVSS 4.0 E:P modifier, elevating real-world urgency despite the authentication prerequisite.

Technical ContextAI

UpTrain is an open-source Python-based platform for evaluating and improving generative AI applications, deployed primarily via Docker. The vulnerable component is the dashboard backend API at uptrain/dashboard/backend/app.py, specifically the /create_project endpoint around line 691. The root cause is CWE-74 (Improper Neutralization of Special Elements in Output Used by a Downstream Component - Injection), which in Python backends typically manifests as user-supplied data passed unsanitized to dynamic execution primitives such as eval(), exec(), or pickle.loads(). The checks and metadata parameters represent structured input fields that the backend processes programmatically, and the lack of input validation allows an attacker to inject executable constructs. No CPE string was provided in the available intelligence, but the affected product is clearly scoped to the UpTrain dashboard backend at version 0.7.1 and below.

RemediationAI

No vendor-released patch has been identified at time of analysis - monitor the GitHub advisory at https://securitylab.github.com/advisories/GHSL-2024-198_GHSL-2024-199_Uptrain/ and the upstream repository for a fix release. As an immediate compensating control, restrict network access to the UpTrain dashboard to trusted internal networks or VPN-only segments, preventing untrusted users from reaching the /create_project endpoint entirely; the trade-off is reduced accessibility for remote teams. If project creation is not actively required, consider disabling or blocking the /create_project route at the reverse proxy or API gateway layer. Review and tighten Docker container configuration: remove privileged mode, restrict host volume mounts, and apply seccomp/AppArmor profiles to limit syscall exposure in the event of exploitation. Enforce strict access control on UpTrain credentials to ensure only trusted personnel hold valid authentication, reducing the pool of potential threat actors.