CVE-2025-6514

CriticalRCE

Published: 09 July 2025

Published

09 July 2025

Modified

15 April 2026

KEV Added

—

Patch

—

CVSS Score v3.1 9.6 CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H

EPSS Score 0.1217 94.0th percentile

Risk Priority 27 60% EPSS · 20% KEV · 20% CVSS

Summary

CVE-2025-6514 is a critical-severity OS Command Injection (CWE-78) vulnerability in Jfrog (inferred from references). Its CVSS base score is 9.6 (Critical).

Operationally, exploitation aligns with the MITRE ATT&CK technique Exploitation for Client Execution (T1203); ranked in the top 6.0% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog.

This vulnerability is AI-related — categorised as AI Agent Protocols and Integrations; in the Protocol-Specific Risks risk domain.

The strongest mitigations our analysis identified are NIST 800-53 SI-10 (Information Input Validation) and SI-2 (Flaw Remediation).

Deeper analysis

CVE-2025-6514 is an OS command injection vulnerability (CWE-78) in mcp-remote that arises when the client processes a crafted authorization_endpoint response URL returned by an untrusted MCP server. The flaw received a CVSS 3.1 score of 9.6, reflecting network attack vector, low complexity, no required privileges, and high impact across confidentiality, integrity, and availability with changed scope.

An unauthenticated remote attacker can exploit the issue by hosting a malicious MCP server that supplies malicious input in the authorization_endpoint field; when a user connects to that server, the injected commands execute on the client system, enabling full remote code execution.

A patch addressing the command-injection vector was merged in commit 607b226a356cb61a239ffaba2fb3db1c9dea4bac of the mcp-remote repository, and JFrog has published detailed analysis and mitigation guidance.

The EPSS score rose to a peak of 0.1606 before settling at the current value of 0.1217, indicating measurable post-disclosure exploitation interest.

EU & UK References

🇪🇺 ENISA EUVD: EUVD-2025-20823

Vulnerability details

mcp-remote is exposed to OS command injection when connecting to untrusted MCP servers due to crafted input from the authorization_endpoint response URL

CWE(s): CWE-78

AI Security AnalysisAI

AI Category: AI Agent Protocols and Integrations
Risk Domain: Protocol-Specific Risks
OWASP Top 10 for LLMs 2025: None mapped
Classification Reason: Matched keywords: mcp

Related Threats

MITRE ATT&CK Enterprise TechniquesAI

T1203 Exploitation for Client Execution Execution

Adversaries may exploit software vulnerabilities in client applications to execute code.

attack.mitre.org →

T1059 Command and Scripting Interpreter Execution

Adversaries may abuse command and script interpreters to execute commands, scripts, or binaries.

attack.mitre.org →

Why these techniques?

OS command injection directly enables arbitrary command execution (T1059) on the client host after user-initiated connection to a malicious server (T1203 Exploitation for Client Execution).

Confidence: HIGH · MITRE ATT&CK Enterprise v18.1