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:
- Cisco admits corporate network compromised by gang with links to Lapsus$
August 11, 2022
Cisco disclosed on Wednesday that its corporate network was accessed by cyber-criminals in May after an employee’s personal Google account was compromised – an act a ransomware gang named “Yanluowang” has now claimed as its work. The world’s largest networking vendor disclosed the months-old compromise after a list of files accessed during the incident appeared on ...
- #StopRansomware: Zeppelin Ransomware
August 11, 2022
The Federal Bureau of Investigation (FBI) and the Cybersecurity and Infrastructure Security Agency (CISA) are releasing this joint CSA to disseminate known Zeppelin ransomware IOCs and TTPs associated with ransomware variants identified through FBI investigations as recently as 21 June 2022. The FBI and CISA encourage organizations to implement the recommendations in the Mitigations section ...
- Automotive supplier breached by 3 ransomware gangs in 2 weeks
August 10, 2022
An automotive supplier had its systems breached and files encrypted by three different ransomware gangs over two weeks in May, two of the attacks happening within just two hours. The attacks followed an initial breach of the company’s systems by a likely initial access broker (IAB) in December 2021, who exploited a firewall misconfiguration to breach ...
- Phishers who breached Twilio and fooled Cloudflare could easily get you, too
August 10, 2022
At least two security-sensitive companies—Twilio and Cloudflare—were targeted in a phishing attack by an advanced threat actor who had possession of home phone numbers of not just employees but employees’ family members as well. In the case of Twilio, a San Francisco-based provider of two-factor authentication and communication services, the unknown hackers succeeded in phishing the ...
- Cisco Talos shares insights related to recent cyber attack on Cisco
August 10, 2022
On May 24, 2022, Cisco became aware of a potential compromise. Since that point, Cisco Security Incident Response (CSIRT) and Cisco Talos have been working to remediate. During the investigation, it was determined that a Cisco employee’s credentials were compromised after an attacker gained control of a personal Google account where credentials saved in the victim’s ...
- APIC fail: Intel ‘Sunny Cove’ chips with SGX spill secrets
August 9, 2022
A group of computer scientists has identified an architectural error in certain recent Intel CPUs that can be abused to expose SGX enclave data like private encryption keys. They call it ÆPIC Leak because it affects the memory-mapped registers of the local Advanced Programmable Interrupt Controller (APIC), which helps the CPU handle interrupt requests from various ...