{
  "id": "N-MACHINE-CAPABILITY-v1.1",
  "type": "report",
  "title": "The tool was not the problem.",
  "author": [
    {
      "family": "Hay",
      "given": "Chris"
    }
  ],
  "issued": {
    "date-parts": [
      [
        2026,
        9,
        12
      ]
    ]
  },
  "URL": "https://chrishayuk.com/records/N-MACHINE-CAPABILITY/1.1",
  "publisher": "Chris Hay",
  "version": "1.1",
  "abstract": "Eighteen assigned AI agents used six web mechanisms at LLM Wilds. All completed the task, including three WebSocket clients and three WebAssembly executions, refuting the predicted drop-off for mechanisms requiring code. The study raises a question about agentic discoverability; it did not test discovery from a site's front door. One harness cache leak also showed how the web an agent receives can differ from the web being served."
}