{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreifoxlh2yabe7z7soi3scolz5ij24i65eyrx5hjq2ehaunr4h5gaju",
"uri": "at://did:plc:4n6wgsqsqm6q2hjncgwmreey/app.bsky.feed.post/3mjpa7vxgd6x2"
},
"coverImage": {
"$type": "blob",
"ref": {
"$link": "bafkreifofzlqr5htxn3nzr7tt4tsiqz6k6dhlkog34wm7kxcbjuiviz5gu"
},
"mimeType": "image/jpeg",
"size": 14091
},
"path": "/post/48949824",
"publishedAt": "2026-04-17T08:30:19.000Z",
"site": "https://programming.dev",
"tags": [
"Programming",
"rimu",
"5 comments",
"https://www.theregister.com/2026/04/16/anthropic_mcp_design_flaw/"
],
"textContent": "submitted by rimu to programming\n70 points | 5 comments\nhttps://www.theregister.com/2026/04/16/anthropic_mcp_design_flaw/\n\nA design flaw – or expected behavior based on a bad design choice, depending on who is telling the story – baked into Anthropic’s official Model Context Protocol (MCP) puts as many as 200,000 servers at risk of complete takeover, according to security researchers.",
"title": "MCP 'design flaw' puts 200k servers at risk and Anthropic won't fix it"
}