MCP tools are implicated in several new attack techniques. Here's a look at how they can be manipulated for good, such as logging tool usage and filtering unauthorized commands.
Over the last few months, there has been a lot of activity in the Model Context Protocol (MCP) space, both in terms of adoption as well as security. Developed by Anthropic, MCP has been rapidly gaining traction across the AI ecosystem. MCP allows Large Language Models (LLMs) to interface with tools and for those interfaces to be rapidly created. MCP tools allow for the rapid development of “agentic” systems, or AI systems that autonomously perform tasks.
Beyond adoption, new attack techniques have been shown to allow prompt injection via MCP tool descriptions and responses, MCP tool poisoning, rug pulls and more.
Prompt Injection is a weakness in LLMs that can be used to elicit unintended behavior, circumvent safeguards and produce potentially malicious responses. Prompt injection occurs when an attacker instructs the LLM to disregard other rules and do the attacker’s bidding. In this blog, I show how to use techniques similar to prompt injection to change the LLM’s interaction with MCP tools. Anyone conducting MCP research may find these techniques useful.
Microsoft synchronization capabilities for managing identities in hybrid environments are not without their risks. In this blog, Tenable Research explores how potential weaknesses in these synchronization options can be exploited.
Synchronizing identity accounts between Microsoft Active Directory (AD) and Entra ID is important for user experience, as it seamlessly synchronizes user identities, credentials and groups between on-premises and cloud-based systems. At the same time, Tenable Research shows the following synchronization options can introduce cybersecurity risk that extend beyond hybrid tenants:
the already known Directory Synchronization Accounts Entra role
the new On Premises Directory Sync Account Entra role
the new Microsoft Entra AD Synchronization Service application
In 2024, Microsoft introduced two new security hardening measures for hybrid Entra ID synchronization. However, despite these improvements, both the Directory Synchronization Accounts and the new On Premises Directory Sync Account roles retain access to critical synchronization APIs. Moreover, the new 'Microsoft Entra AD Synchronization Service' application exposes the privileged ADSynchronization.ReadWrite.All permission, introducing another potential attack path that security teams must watch closely.
In this technical blog, we break down the changes Microsoft made to each of its synchronization options, explore where new risks were introduced and provide guidance on how Tenable Identity Exposure can help you monitor and secure your hybrid synchronization environment.
Salt Typhoon, a state-sponsored actor linked to the People’s Republic of China, has breached at least nine U.S.-based telecommunications companies with the intent to target high profile government and political figures. Tenable Research examines the tactics, techniques and procedures of this threat actor.
A security issue exists in D-Link D-View 8 v2.0.2.89 and prior that could allow an attacker to manipulate the probe inventory of the D-View service. This could result in the disclosure of info
A researcher at Tenable has discovered an issue that enables limited, unauthorized access to cross-tenant applications and sensitive data (including but not limited to authentication secrets). Background The issue occurred as a result of insufficient access control to Azure Function hosts, which are launched as part of the creation and operation of custom connectors in Microsoft’s Power Platform (Power Apps, Power Automation).
Last week, Senator Ron Wyden sent a letter to the Cybersecurity and Infrastructure Security Agency (CISA), the Department of Justice and the Federal Trade Commission (FTC) asking that they hold Microsoft accountable for a repeated pattern of negligent cybersecurity practices, which has enabled Chine
Paid Memberships Pro : CVE-2023-23488 - Unauthenticated SQL Injection
Easy Digital Downloads: CVE-2023-23489 - Unauthenticated SQL Injection
Survey Maker: CVE-2023-23490 - Authenticated SQL Injection