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:
- U.S. arm of China mega-lender ICBC hit by ransomware attack
November 10, 2023
The U.S. arm of China’s largest bank said Thursday that it was hit by a ransomware attack, forcing clients to reroute trades and disrupting the U.S. Treasury market. Ransomware attacks typically access vulnerable computer systems and encrypt or steal data, before sending a ransom note demanding payment in exchange for decrypting the data or not releasing ...
- Modern Asian APT groups’ tactics, techniques and procedures (TTPs)
November 9, 2023
This report consists of six main sections – Incidents involving Asian APT groups in various regions of the planet Information on five unique incidents that Kaspersky researchers detected in different parts of the world. Each incident is a unique case within a specific country and industry, and they provide a description of the actions and TTPs ...
- Secretary General: Through NATO, we can build a secure cyberspace for all
November 9, 2023
The Secretary General emphasized that cyber is driving strategic competition and that authoritarian regimes, including China and Russia, are: “challenging our interests, our values and our security.” He said they are: “determined to shape the future of cyberspace in own image with little transparency and no regard for human rights.” At the Vilnius Summit this ...
- Threat Predictions for 2024: Chained AI and CaaS Operations Give Attackers More “Easy” Buttons Than Ever
November 9, 2023
With the growth of Cybercrime-as-a-Service (CaaS) operations and the advent of generative AI, threat actors have more “easy” buttons at their fingertips to assist with carrying out attacks than ever before. By relying on the growing capabilities in their respective toolboxes, adversaries will increase the sophistication of their activities. They’ll launch more targeted and stealthier hacks ...
- Law firm Allen & Overy hit by ‘data incident’
November 9, 2023
Allen & Overy has suffered a “data incident”, the London-founded law firm said on Thursday, after social media posts suggested it had been hacked by the Lockbit cybercrime gang. The attack, first reported by the Financial Times, comes after seven countries, including the U.S. and Britain, in June named Lockbit as the world’s top ransomware threat. Read ...
- OpenAI Blames ChatGPT’s Intermittent Outages On ‘Abnormal Traffic’ That Suggests Potential Cyber Attack
November 9, 2023
ChatGPT continued to face intermittent outages late Wednesday, which the platform’s maker OpenAI blamed on a potential cyberattack, hours after the AI chatbot platform recovered from a wide outage that the company initially attributed to a surge in interest for its new features. Early on Thursday, OpenAI’s service status page displayed a notification saying both ChatGPT ...