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:
- Vice Society Ransomware Group Targets Manufacturing Companies
January 24, 2023
The Vice Society ransomware group made headlines in late 2022 and early 2023 during a spate of attacks against several targets, such as the one that affected the rapid transit system in San Francisco. Most reports have the threat actor focusing its efforts on the education and the healthcare industries. However, through Trend Micro’s telemetry data, ...
- New wave of attacks use ProxyNotShell/OWASSRF vulnerabilities to target Microsoft Exchange
January 24, 2023
Researchers at S.C. Bitdefender SRL today warned of a new wave of attacks using known vulnerabilities to target Microsoft Exchange. The researchers started to notice an increase in attacks using ProxyNotShell/OWASSRF exploits to target on-premises Microsoft Exchange deployments at the end of November. The Server-Side Request Forgery attacks allow an attacker to send a crafted request ...
- LastPass owner GoTo says hackers stole customers’ backups
January 24, 2023
LastPass’ parent company GoTo – formerly LogMeIn – has confirmed that cybercriminals stole customers’ encrypted backups during a recent breach of its systems. The breach was first confirmed by LastPass on November 30. At the time, LastPass chief executive Karim Toubba said an “unauthorized party” had gained access to some customers’ information stored in a third-party ...
- Hackers use Golang source code interpreter to evade detection
January 24, 2023
A Chinese-speaking hacking group tracked as ‘DragonSpark’ was observed employing Golang source code interpretation to evade detection while launching espionage attacks against organizations in East Asia. The attacks are tracked by SentinelLabs, whose researchers report that DragonSpark relies on a little-known open-source tool called SparkRAT to steal sensitive data from compromised systems, execute commands, perform lateral ...
- Apple fixes actively exploited iOS zero-day on older iPhones, iPads
January 23, 2023
Apple has backported security patches addressing a remotely exploitable zero-day vulnerability to older iPhones and iPads. This bug is tracked as CVE-2022-42856, and it stems from a type confusion weakness in Apple’s Webkit web browser browsing engine. Read more… Source: Bleeping Computer
- FBI Confirms Lazarus Group Cyber Actors Responsible for Harmony’s Horizon Bridge Currency Theft
January 23, 2023
The FBI continues to combat malicious cyber activity, including the threat posed by the Democratic People’s Republic of Korea (DPRK) to the U.S. and our private sector partners. Through our investigation, we were able to confirm that the Lazarus Group (also known as APT38), cyber actors associated with the DPRK, are responsible for the theft ...