reloadium.com

showConnectionMatrix

Show a cross-reference matrix between two sets of things — rows and columns — where each cell is a confidence score (0-1) with an optional reason, for 'how does each of these relate to each of those, and why' questions (entities against events in an investigation, a candidate's skills against a job's requirements, sources against the claims they support, debate arguments against rebuttals). Not for ranking options to pick a winner — use showComparisonMatrix for that instead; this tool has no notion of a winner or a weighted total, it's a read-only cross-reference. Unlike showComparisonMatrix, cells are SPARSE — omit a row/column pair entirely when there's no real connection, don't invent a low-confidence cell just to fill the grid; most real cross-reference matrices are mostly empty, and a dense one usually means the axes weren't independent enough. Provide `rows` and `cols`, each an array of {id, name, subtitle?} — subtitle is a short secondary line (e.g. a person's role, an event's date, a requirement's category) — and `cells`, an array of only the pairs that actually connect: {rowId, colId, confidence: 0-1, reason?}. Confidence is how strong/certain the connection is, not how important it is — a well-evidenced minor link can be high confidence, a speculative major one low. Give a `reason` for any cell whose confidence is worth explaining; a bare number with no reason is a weak signal on its own and won't appear in the detail list the full view shows. Optionally set `rowsLabel`/`colsLabel` (e.g. 'Entities'/'Events', 'Skills'/'Requirements') to name the two axes — defaults to 'Rows'/'Columns'.

Page: /

Input schema

{
  "type": "object",
  "properties": {
    "title": {
      "type": "string"
    },
    "rowsLabel": {
      "type": "string"
    },
    "colsLabel": {
      "type": "string"
    },
    "rows": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "id": {
            "type": "string"
          },
          "name": {
            "type": "string"
          },
          "subtitle": {
            "type": "string"
          }
        },
        "required": [
          "id",
          "name"
        ]
      }
    },
    "cols": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "id": {
            "type": "string"
          },
          "name": {
            "type": "string"
          },
          "subtitle": {
            "type": "string"
          }
        },
        "required": [
          "id",
          "name"
        ]
      }
    },
    "cells": {
      "type": "array",
      "description": "Only the row/column pairs that actually connect — omit the rest.",
      "items": {
        "type": "object",
        "properties": {
          "rowId": {
            "type": "string"
          },
          "colId": {
            "type": "string"
          },
          "confidence": {
            "type": "number",
            "description": "0-1."
          },
          "reason": {
            "type": "string"
          }
        },
        "required": [
          "rowId",
          "colId",
          "confidence"
        ]
      }
    }
  },
  "required": [
    "rows",
    "cols",
    "cells"
  ]
}