{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreicwnz723mzrppsub3urjphahitxl4p7xnu74dkid27jucywfkx3ou",
    "uri": "at://did:plc:xb5etev7ncohgebas3kjeqek/app.bsky.feed.post/3mfwmyltzppd2"
  },
  "coverImage": {
    "$type": "blob",
    "ref": {
      "$link": "bafkreiaqmw6kjlnjm3y4m4u2gqy5rqr3lmslmujpz67wj4smifveof3ehy"
    },
    "mimeType": "image/jpeg",
    "size": 20385
  },
  "path": "/openclaw-vulnerability-openclaw-hijack-ai-agents/",
  "publishedAt": "2026-02-27T16:39:06.000Z",
  "site": "https://hackread.com",
  "tags": [
    "Security",
    "Artificial Intelligence",
    "AI Agents",
    "ClawJacked",
    "Cybersecurity",
    "OpenClaw",
    "Peter Steinberger",
    "Privacy",
    "Vulnerability"
  ],
  "textContent": "Is your AI assistant safe? Oasis Security researchers have found a critical ClawJacked vulnerability in OpenClaw that allows hackers to hijack AI agents through a simple browser tab.",
  "title": "ClawJacked Vulnerability in OpenClaw Could Let Websites Hijack AI Agents"
}