Skip to content
Threat Feed

Product

Claude Code

5 briefs RSS
high advisory

AI Agent Data Theft via Indirect Prompt Injection

Attackers are leveraging indirect prompt injection against AI agents with access to private data, untrusted content, and external communication channels to steal sensitive information by embedding malicious instructions in content processed by the agent.

GitHub Copilot Agent +4 ai-agent prompt-injection data-theft ai-security
1r 2t
critical threat

Shai-Hulud Malware Used in Supply Chain Attack via Compromised npm Packages

The Shai-Hulud malware was used in a large-scale software supply-chain attack compromising hundreds of packages across open-source software ecosystems by compromising developer secrets and CI/CD pipelines.

router +11 TeamPCP supply-chain supply-chain-attack npm pypi credential-theft shai-hulud
3r 7t 3i
critical advisory

Claude Code OAuth Token Theft via MCP Hijacking

Attackers can silently redirect Claude Code MCP traffic to intercept OAuth tokens, enabling persistent access to connected SaaS platforms by modifying the ~/.claude.json file in a man-in-the-middle attack.

Claude Code oauth man-in-the-middle credential access
3r 2t
critical advisory

AI Coding Agents Vulnerable to Supply Chain Attacks via Malicious Repositories

AI coding agents like Claude Code, Gemini CLI, Cursor CLI, and GitHub Copilot Agents can be manipulated to introduce malicious code into software supply chains by accessing attacker-controlled repositories, leading to potential remote code execution and supply chain compromises.

Claude Code +3 supply chain ai remote code execution
2r 1t
high advisory

Claude Code Trust Dialog Bypass via Git Worktree Spoofing

A vulnerability in Claude Code allowed for trust dialog bypass via git worktree spoofing, potentially leading to arbitrary code execution by crafting a malicious repository with a `commondir` file pointing to a previously trusted path, bypassing the trust dialog, and executing malicious hooks defined in `.claude/settings.json`.

Claude Code git code-execution trust-bypass
2r 1t