Skip to content
Threat Feed
high threat

Cursor: Supply chain compromise via Fireworks AI breach

Attackers leveraged a compromised access key from Fireworks AI to gain unauthorized access to Cursor internal systems and inject malicious copies of the Anthropic Claude Code extension into the Cursor VS Code marketplace cache.

exploited anthropic.claude-code extension
3t
medium advisory

Suspicious IDE Extension Installation via VSIX

Threat actors can gain persistence or code execution by sideloading malicious IDE extensions via the VSIX format, often triggered by script hosts or from volatile staging directories.

Visual Studio Code +3 persistence ide-extension software-supply-chain endpoint
high advisory

Anthropic Claude VS Code extension malicious update distribution

An unknown actor compromised an official Anthropic distribution channel to push a malicious version of the Claude VS Code extension, enabling persistent SSH backdoors and credential theft on developer workstations.

exploited Claude Code Extension +3 supply-chain ide trojan persistence exfiltration backdoor credential-theft developer-tools +2
11t updated
high advisory

VS Code and Cursor: Credential harvesting by IDE extensions

Malicious extensions or compromised IDE agents may use VS Code or Cursor process trees to exfiltrate multiple sensitive credential stores from a developer workstation.

Visual Studio Code +1 credential-access persistence ide supply-chain
2t
critical advisory

Suspicious Activity Detection from GenAI Coding Utilities

This detection rule identifies suspicious endpoint activity, such as malicious file creation or shellcode execution, originating from or triggered by AI-assisted coding and assistant tools indicating potential supply chain or prompt injection abuse.

Cursor +11 supply-chain endpoint-security llm-security unauthorized-ai-usage
1r 1t updated
medium advisory

GenAI Tool Configuration Poisoning via MCP Server Injection

Adversaries are targeting configuration files of popular GenAI tools to inject malicious Model Context Protocol (MCP) servers, enabling persistence, arbitrary command execution, and data exfiltration.

Cursor +9 defense-evasion persistence genai-security supply-chain
1r 2t
high threat

Arbitrary Code Execution in AI Coding Agents via Git Configuration

Multiple AI coding agents are vulnerable to arbitrary code execution due to the automated, unsandboxed execution of commands defined within a repository's local Git configuration, specifically the 'core.fsmonitor' setting.

exploited goose +8 supply-chain rce ai-security git
1r 1t
medium threat

Detection of Generative AI Processes Connecting to Unusual Domains

Adversaries may compromise macOS-based Generative AI (GenAI) tools through prompt injection, malicious Model Context Protocol (MCP) servers, or poisoned plugins to establish Command and Control (C2) channels or exfiltrate sensitive data by causing them to connect to unusual domains.

exploited Claude +8 command-and-control genai macos data-exfiltration
1r 1t
high advisory

CVE-2026-63093: Binary Planting Vulnerability in Cursor for Windows

CVE-2026-63093 describes a binary planting vulnerability in Cursor for Windows version 3.2.16 that allows a remote attacker to achieve arbitrary code execution by placing a malicious `git.exe` file in a crafted repository's root, which the Cursor IDE automatically executes during startup or on a recurring cadence when a developer opens the repository, running the malicious binary under the privileges of the current user.

Cursor for Windows 3.2.16 binary-planting ide vulnerability execution windows
1r 3t 1c
high advisory

Security Risks Associated with AI Coding Tools, Including GhostApproval Vulnerability

The adoption of AI coding tools introduces significant security risks, such as the generation of vulnerable code with OWASP Top 10 flaws, the inadvertent leakage of sensitive secrets and hardcoded credentials, and supply chain compromise via 'slopsquatting,' alongside specific vulnerabilities like 'GhostApproval' which allows remote code execution on developer machines.

ai coding-tools supply-chain vulnerability rce credential-exposure
3t
high advisory

Shai-Hulud Campaign Activity

Tracking brief for the Shai-Hulud campaign; individual sightings are folded in as reported.

jscrambler 8.14.0 +102 campaign shai-hulud
20i updated
critical advisory

Elastic Defend Alert from GenAI Utility or Descendant

This rule detects Elastic Defend alerts originating from or directly related to GenAI coding utilities, indicating potential prompt injection, malicious skills, or supply-chain compromise.

Cursor +11 genai supply-chain elastic-defend
2r 1t
critical advisory

Unsecured Model Context Protocol (MCP) Server Deployments Expose AI Integrations

Unsecured Model Context Protocol (MCP) servers, used to connect AI agents to enterprise tools, lack authentication and audit trails, leading to data exfiltration, private repo leaks, cross-tenant exposure, and remote code execution due to AI agents using valid user credentials to make API calls based on potentially poisoned context.

Claude +4 ai mcp zero-trust data-exfiltration rce
2r 8t 1i
medium advisory

Unusual Modification of GenAI Tool Configuration File

This rule detects unusual modifications to GenAI tool configuration files, potentially indicating an attacker injecting malicious MCP server configurations to hijack AI agents for persistence, command and control, or data exfiltration.

Claude +3 genai configuration-modification defense-evasion
3r 4t