{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreiekoydwbj77jut7hm62pvtqsevyyzo4ktgvtfbb7r2u3hjcdtrvra",
"uri": "at://did:plc:ghkvexthfanuyq7fb5veq6tw/app.bsky.feed.post/3mpjsmf4dhoj2"
},
"coverImage": {
"$type": "blob",
"ref": {
"$link": "bafkreievdadpixz3thfbk3ppxoezwhsuwbjnwrzsl4h45nl6bouftwoopy"
},
"mimeType": "image/jpeg",
"size": 284497
},
"path": "/2026/06/microsoft-warns-poisoned-mcp-tool.html",
"publishedAt": "2026-06-30T17:46:07.000Z",
"site": "https://thehackernews.com",
"textContent": "New Microsoft research shows how attackers can hijack AI agents that act on a user's behalf, using nothing more than a poisoned tool description to make the agent quietly hand over company data to an outsider. The trick is that the agent never breaks a rule. Every step looks routine, so in a default setup no alarm may fire. The work comes from Microsoft Incident Response and its",
"title": "Microsoft Warns Poisoned MCP Tool Descriptions Can Make AI Agents Leak Data"
}