CVE-2026-123019
Deferred Deferred - Pending Action

Remote Code Execution in eosphoros-ai DB-GPT

Publication date: 2026-11-05

Last updated on: 2026-11-05

Assigner: VulDB

Description
A weakness has been identified in eosphoros-ai DB-GPT up to 0.8.2. Impacted is the function exec_module of the file hf_adapter.py of the component HuggingFace Adapter. Executing a manipulation of the argument trust_remote_code can lead to inclusion of functionality from untrusted control sphere. The attack can be launched remotely. The exploit has been made available to the public and could be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-11-05
Last Modified
2026-11-05
Generated
2026-11-05
AI Q&A
2026-11-05
EPSS Evaluated
N/A
NVD
CVE-2026-123019
EUVD
EUVD-2026-117415

Affected Vendors & Products

Showing 3 associated CPEs
Vendor Product Version / Range
eosphoros-ai DB-GPT 0.8.0
eosphoros-ai DB-GPT 0.8.1
eosphoros-ai DB-GPT 0.8.2

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-829 The product imports, requires, or includes executable functionality (such as a library) from a source that is outside of the intended control sphere.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

CVE-2026-123019 is a critical vulnerability in DB-GPT versions up to 0.8.2 where the HuggingFace Adapter's exec_module function in hf_adapter.py allows manipulation of the trust_remote_code argument. This enables inclusion of untrusted code from remote sources, leading to remote code execution. The attack is remotely exploitable without authentication and a public exploit exists.

Detection Guidance

Check if DB-GPT is running on port 5670 by default. Inspect network traffic for unauthenticated POST requests to /api/v2/serve/model/models. Look for unexpected model registrations or code execution markers in logs. Verify if trust_remote_code is set to True in hf_adapter.py or HFLLMDeployModelParameters.

Impact Analysis

An attacker could exploit this to execute arbitrary code on your DB-GPT server, gaining full control over the system. This includes accessing sensitive data like database credentials, environment variables, or writing arbitrary files. The attack requires only network access to the default port 5670 and works without Docker or prior configuration.

Compliance Impact

This vulnerability likely violates compliance requirements for data protection and security. It enables unauthorized access to sensitive data (potentially including personal health or financial information), which could result in violations of GDPR, HIPAA, or other regulations. The lack of authentication and remote code execution capability creates significant compliance risks.

Mitigation Strategies

Disable trust_remote_code by setting it to False in all relevant configurations. Add authentication to the model registration endpoint. Restrict network access to DB-GPT's default port 5670. Monitor for unauthorized model registrations or suspicious API calls.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-123019. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart

Security Radar

News mentioning this vulnerability
Loading related security news...