Cyber Posture

CVE-2025-0185

HighPublic PoC

Published: 20 March 2025

Published
20 March 2025
Modified
27 March 2025
KEV Added
Patch
CVSS Score 8.8 CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
EPSS Score 0.0302 86.7th percentile
Risk Priority 19 60% EPSS · 20% KEV · 20% CVSS

Description

Adversaries may abuse Python commands and scripts for execution.

Security Summary

CVE-2025-0185 is a Pandas Query Injection vulnerability in the Dify Tools' Vanna module of the langgenius/dify repository, affecting the latest version. The flaw occurs in the function `vn.get_training_plan_generic(df_information_schema)`, which does not properly sanitize user inputs before executing queries via the Pandas library. Published on 2025-03-20, this issue is classified under CWE-94 (Code Injection) with a CVSS v3.1 base score of 8.8 (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), indicating high severity due to potential remote code execution (RCE).

The vulnerability can be exploited by low-privileged remote attackers (PR:L) with network access (AV:N), requiring low attack complexity (AC:L) and no user interaction (UI:N). Exploitation involves injecting malicious queries through unsanitized inputs, enabling attackers to execute arbitrary code on the affected system and achieve high impacts on confidentiality, integrity, and availability (C:H/I:H/A:H).

Advisories and mitigation guidance are detailed in the Huntr bounty report at https://huntr.com/bounties/7d9eb9b2-7b86-45ed-89bd-276c1350db7e.

Details

CWE(s)
CWE-94

Affected Products

dify
dify
all versions

AI Security Analysis

AI Category
Enterprise AI Assistants
Risk Domain
Data-Related Vulnerabilities
OWASP Top 10 for LLMs 2025
None mapped
MITRE ATLAS Techniques
None mapped
Classification Reason
Dify (langgenius/dify) is an open-source enterprise platform for developing LLM applications and AI assistants, with the Vanna module enabling AI-powered text-to-SQL functionality; the vulnerability in its training plan generation is AI-related.

MITRE ATT&CK Enterprise Techniques

T1190 Exploit Public-Facing Application Initial Access
Adversaries may attempt to exploit a weakness in an Internet-facing host or system to initially access a network.
T1059.006 Python Execution
Adversaries may abuse Python commands and scripts for execution.
Why these techniques?

Pandas Query Injection vulnerability in Dify's Vanna module enables remote code execution via unsanitized inputs, exploiting public-facing applications (T1190) and facilitating arbitrary Python command execution (T1059.006).

References