This article examines the security implications of the Model Context Protocol (MCP) sampling feature in the context of a widely used coding copilot application.
MCP is a standard for connecting large language model (LLM) applications to external data sources and tools. We show that, without proper safeguards, malicious MCP servers can exploit the sampling feature for a range of attacks. We demonstrate these risks in practice through three proof-of-concept (PoC) examples conducted within the coding copilot, and discuss strategies for effective prevention.
Read more…
Source: Palo Alto Unit 42
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- Threat Actors Chaining Unpatched VMware Vulnerabilities for Full System Control
June 2, 2022
The Cybersecurity and Infrastructure Security Agency (CISA) is releasing this CSA to warn organizations that malicious cyber actors, likely advanced persistent threat (APT) actors, are exploiting CVE-2022-22954 and CVE-2022-22960 separately and in combination. These vulnerabilities affect certain versions of VMware Workspace ONE Access, VMware Identity Manager (vIDM), VMware vRealize Automation (vRA), VMware Cloud Foundation, and ...
- YourCyanide: A CMD-Based Ransomware With Multiple Layers of Obfuscation
June 2, 2022
The Trend Micro Threat Hunting team recently analyzed a series of CMD-based ransomware variants with a number capabilities such as stealing user information, bypassing remote desktop connections, and propagating through email and physical drives. In this blog entry, Trend Micro researchers will analyze YourCyanide, the latest variant of the CMD-based ransomware family that started with GonnaCope. ...
- Takedown of SMS-based FluBot spyware infecting Android phones
June 1, 2022
An international law enforcement operation involving 11 countries has resulted in the takedown of one of the fastest-spreading mobile malware to date. Known as FluBot, this Android malware has been spreading aggressively through SMS, stealing passwords, online banking details and other sensitive information from infected smartphones across the world. Its infrastructure was successfully disrupted earlier ...
- Using Python to unearth a goldmine of threat intelligence from leaked chat logs
June 1, 2022
Dealing with a great amount of data can be time consuming, thus using Python can be very powerful to help analysts sort information and extract the most relevant data for their investigation. The open-source tools library, MSTICPy, for example, is a Python tool dedicated to threat intelligence. It aims to help threat analysts acquire, enrich, ...
- FBI: Karakurt Data Extortion Group
June 1, 2022
The Federal Bureau of Investigation (FBI), the Cybersecurity and Infrastructure Security Agency (CISA), the Department of the Treasury (Treasury), and the Financial Crimes Enforcement Network (FinCEN) are releasing this joint Cybersecurity Advisory (CSA) to provide information on the Karakurt data extortion group, also known as the Karakurt Team and Karakurt Lair. Karakurt actors have employed a ...
- Patch Your WSO2: CVE-2022-29464 Exploited to Install Linux-Compatible Cobalt Strike Beacons, Other Malware
May 31, 2022
Trend Micro researchers observed vulnerability CVE-2022-29464 being exploited in the wild since April, allowing unrestricted file uploads resulting to arbitrary remote code execution (RCE). Disclosed and patched in April, the security gap was ranked Critical at 9.8 and affects a number of WSO2 products. It requires no user interaction and administrative privileges for abuse, and ...