Skip to content

28 eval-informed mental models and critical-thinking skills for Claude Code, GitHub Copilot, Codex, Cursor, and other Agent Skills-compatible tools

606 B
Raw Blame
{
  "stack": [
    "first-principles",
    "second-order",
    "red-team"
  ],
  "solver": "claude-sonnet-4-6",
  "problems": 9,
  "stack:placebo": {
    "wins": 5,
    "losses": 0,
    "ties": 4,
    "n": 9,
    "decisive": 5,
    "win_rate": 0.778,
    "ci95": [
      0.566,
      1
    ],
    "p_value": 0.074,
    "significant": false,
    "powered": true
  },
  "stack:single": {
    "wins": 5,
    "losses": 1,
    "ties": 3,
    "n": 9,
    "decisive": 6,
    "win_rate": 0.722,
    "ci95": [
      0.436,
      0.97
    ],
    "p_value": 0.221,
    "significant": false,
    "powered": false
  }
}