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:
- PyPI warns of first-ever phishing campaign against its users
August 26, 2022
The Python Package Index, better known among developers as PyPI, has issued a warning about a phishing attack targeting developers who use the service. The community-run organization said this is the first known phishing attack against PyPI users. And the attack has unfortunately been somewhat successful, resulting in the compromise of some users’ accounts. PyPI is an ...
- Cyber criminals are launching phishing attacks on LinkedIn
August 25, 2022
Regular users of LinkedIn, the professional networking and social working platform, have noticed an increase of threat actors trying to steal critical personal information through phishing attacks. These cyber criminals are using false LinkedIn accounts to trick unsuspecting victims into giving up confidential information. How are they doing it? Threat actors start by creating fraudulent LinkedIn ...
- MERCURY leveraging Log4j 2 vulnerabilities in unpatched systems to target Israeli organizations
August 25, 2022
In recent weeks, the Microsoft Threat Intelligence Center (MSTIC) and Microsoft 365 Defender Research Team detected Iran-based threat actor MERCURY leveraging exploitation of Log4j 2 vulnerabilities in SysAid applications against organizations all located in Israel. MSTIC assesses with high confidence that MERCURY’s observed activity was affiliated with Iran’s Ministry of Intelligence and Security (MOIS). While MERCURY ...
- Kimsuky’s GoldDragon cluster and its C2 operations
August 25, 2022
Kimsuky (also known as Thallium, Black Banshee and Velvet Chollima) is a prolific and active threat actor primarily targeting Korea-related entities. Like other sophisticated adversaries, this group also updates its tools very quickly. In early 2022, Kaspersky researchers observed this group was attacking the media and a think-tank in South Korea and reported technical details ...
- LastPass developer systems hacked to steal source code
August 25, 2022
Password management firm LastPass was hacked two weeks ago, enabling threat actors to steal the company’s source code and proprietary technical information. The disclosure comes after BleepingComputer learned of the breach from insiders last week and reached out to the company on August 21st without receiving a response to our questions. Sources told BleepingComputer that employees were ...
- CISA: Preparing Critical Infrastructure for Post-Quantum Cryptography
August 24, 2022
Nation-states and private companies are actively pursuing the capabilities of quantum computers. Quantum computing opens up exciting new possibilities; however, the consequences of this new technology include threats to the current cryptographic standards. These standards ensure data confidentiality and integrity and support key elements of network security. While quantum computing technology capable of breaking public ...