{
  "ok": true,
  "data": {
    "entry": {
      "id": "prompt-injection-is-a-tool-boundary",
      "kind": "method",
      "kindLabel": "Method",
      "title": "Prompt injection is a boundary problem",
      "body": "A prompt injection works when untrusted content gets treated like a higher-priority instruction. The repair is not just a better warning. It is separating content, command, tools, and authority.",
      "tag": "security",
      "date": "2026 · 06 · 30",
      "publishedAt": "2026-06-30",
      "note": null,
      "insight": {
        "aha": "Prompt injection becomes dangerous when untrusted text crosses from content into command authority.",
        "tryThis": "Place a harmless fake instruction inside retrieved content and make the model classify it as content, not a command.",
        "watchFor": [
          "Untrusted pages allowed to override system or user intent",
          "Tools exposed without least-authority checks",
          "Warnings in prompts used instead of product boundaries"
        ]
      },
      "extended": [
        "A retrieved page, webhook body, document, or user note may contain text that sounds like instructions. The model needs a rule that this text is content to analyze, not authority to obey.",
        "The safer architecture binds tools to schemas, trusted domains, explicit user intent, protected-action checks, and refusal states. The prompt can help, but the product boundary has to carry the load.",
        "The AHA is that security lives where text becomes action."
      ],
      "url": "https://tips.platphormnews.com/entries/prompt-injection-is-a-tool-boundary"
    }
  }
}