{
  "id": 18312749,
  "title": "Anthropic won't fix a Claude Dispatch flaw that could let attackers execute remote commands across corporate boundaries",
  "url": "https://urgent.news/2026/11/05/anthropic-wont-fix-a-claude-dispatch-flaw-that-could-let-attackers-18312749",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-05T13:00:00.000Z",
  "source": {
    "name": "The Register",
    "slug": "the-register",
    "url": "https://www.theregister.com/research/2026/11/05/anthropic-wont-fix-a-claude-dispatch-flaw-that-could-let-attackers-execute-remote-commands-across-corporate-boundaries/5307684"
  },
  "original_language": "en",
  "account": "A security flaw exists in Anthropic's Claude Dispatch feature that could allow attackers to execute remote commands across corporate networks, according to Forever Security researcher Gal Weizman. Claude Dispatch enables users to send and manage tasks from mobile devices or browsers to Claude Desktop, an AI agent that runs commands on a computer. Weizman discovered that Claude Dispatch's interface and API could be accessed through a standard web session on claude.ai, enabling a malicious browser extension to issue commands to Claude Desktop as if they came from the legitimate Dispatch user. This could grant attackers access to the operating system's files, enabling them to run code and potentially compromise the machine. Anneal acknowledges the vulnerability but did not address the issue, instead planning to retire Dispatch at the end of October, which has not occurred. The flaw persists in other Claude features like Claude Code Remote Control and Claude Cowork sessions. Weizman demonstrated the attack process in six steps, which involve installing a malicious browser extension, authenticating to both Claude on the web and Claude Desktop, and using prompts to convince Claude Desktop to execute malicious commands.",
  "summary": "From EOP to RCE in six simple steps",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}