Remote Code Execution Vulnerability in MCP Inspector by Model Context Protocol
CVE-2025-49596

9.4CRITICAL

Key Information:

Status
Vendor
CVE Published:
13 June 2025

Badges

📈 Trended📈 Score: 1,740👾 Exploit Exists📰 News Worthy

What is CVE-2025-49596?

CVE-2025-49596 is a significant vulnerability found in the MCP Inspector, a developer tool that facilitates testing and debugging of Model Context Protocol (MCP) servers. The tool is designed to streamline the development process by allowing developers to interact with MCP servers efficiently. However, versions prior to 0.14.1 exhibit a critical security flaw that enables remote code execution through an unauthenticated connection between the Inspector client and its proxy. This lack of authentication means malicious actors can send unauthorized requests to execute arbitrary MCP commands using standard input/output channels (stdio). The ramifications for organizations using this software can be severe, as attackers could misuse the vulnerability to manipulate server operations, inject malicious commands, or gain unauthorized access to sensitive data.

Potential impact of CVE-2025-49596

  1. Remote Code Execution: Exploiting this vulnerability can allow attackers to execute arbitrary commands on the MCP server without authentication, potentially leading to full system control.

  2. Data Compromise: Attackers could leverage the vulnerability to access, modify, or delete sensitive data stored on MCP servers, posing serious risks to data integrity and confidentiality.

  3. Increased Attack Surface: The existence of this vulnerability expands the potential entry points for attackers within an organization's infrastructure, increasing the likelihood of further exploits and enhancing the overall risk profile of the affected systems.

Affected Version(s)

inspector < 0.14.1

News Articles

Anthropic MCP Inspector Tool Vulnerability Let Attackers Execute Arbitrary Code on Developer Machines

A critical Remote Code Execution (RCE) vulnerability in Anthropic's MCP Inspector tool, designated as CVE-2025-49596 with a severe CVSS score of 9.4. 

Anthropic MCP Inspector Vulnerability Lets Hackers Run Arbitrary Code Remotely

A newly disclosed vulnerability in Anthropic’s Model Context Protocol (MCP) Inspector tool has sent shockwaves through the AI development community.

Critical Vulnerability in Anthropic's MCP Exposes Developer Machines to Remote Exploits

Critical RCE vulnerability discovered in Anthropic's MCP Inspector, impacting AI developers and networks.

References

CVSS V4

Score:
9.4
Severity:
CRITICAL
Confidentiality:
High
Integrity:
High
Availability:
High
Attack Vector:
Network
Attack Complexity:
Low
Attack Required:
None
Privileges Required:
Undefined
User Interaction:
Unknown

Timeline

  • 📈

    Vulnerability started trending

  • 👾

    Exploit known to exist

  • 📰

    First article discovered by The Hacker News

  • Vulnerability published

  • Vulnerability Reserved

.