Code Injection in eosphoros-ai DB-GPT CodeAction Component

A security flaw has been discovered in eosphoros-ai DB-GPT up to 0.8.2. This issue affects the function execute_code of the file code_utils.py of the component CodeAction. Performing a manipulation results in code injection. The attack can be initiated remotely. The exploit has been released to the public and may be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.

Casky was already ahead

This CVE exploits attack patterns that Casky's 323matched skills already investigate — long before this vulnerability was disclosed. Claude's reasoning model maps these techniques to MITRE ATT&CK, so practitioners who ran these skills have already seen the threat behaviour in their findings.

Analysis

A critical code injection vulnerability exists in eosphoros-ai DB-GPT versions up to 0.8.2, specifically within the execute_code function of code_utils.py. This flaw allows attackers to inject and execute arbitrary code remotely by manipulating input to the CodeAction component. The vulnerability is particularly concerning because DB-GPT is designed to facilitate AI-driven database operations, meaning compromised instances could provide attackers with direct access to sensitive data stores and backend systems. Organizations deploying DB-GPT for data analysis, automation, or integration workflows are at immediate risk, especially since the exploit has been publicly disclosed and no official patch has been released despite early notification to the project maintainers.

Casky's 323 matched skills leverage Claude's extended reasoning to detect the attack patterns behind this CVE by analyzing both initial access vectors (TA0001) and resource development activities (TA0043). Practitioners using Casky would identify suspicious code patterns in the CodeAction component logs showing unsanitized input flowing into execute_code, along with evidence of dynamic code generation or eval-like operations. The platform's skills map to CWE-74 (Improper Neutralization of Special Elements in Output) and CWE-94 (Improper Control of Generation of Code), enabling detection of malicious payloads attempting to break out of intended execution contexts. Security teams would surface findings showing unauthorized code execution attempts, unexpected subprocess spawning, or database query manipulation originating from the vulnerable code path—allowing them to identify and isolate compromised DB-GPT instances before attackers establish persistence.

Live Threat IntelligenceComing soon

Casky Risk Score

—

EPSS Probability

—

Shodan Exposure

—

GreyNoise Signal

—

Composite risk scoring from EPSS, CISA KEV, Shodan, and GreyNoise — 21 security APIs correlated into a single Casky Risk Score. Coming in Casky Pro. Join early access →

323 Casky skills cover this attack pattern

These skills use Claude AI's reasoning model to surface findings in the same attack categories as CVE-2026-123018.

Abuse of Public-Facing API: Mobile API Abuse

mobile security · low

Run

abusing-dpapi-for-credential-access

red teaming · high

Run

abusing-shadow-credentials-for-privesc

red teaming · high

Run

MITRE ATT&CK Techniques

TA0001 TA0043

Investigate the attack patterns behind CVE-2026-123018

Casky has 323 skills that investigate the attack patterns behind CVE-2026-123018. Run one and get CVSS-scored findings in 3 minutes.

Run the skill that detects this →

X Instagram LinkedIn