Allocate by Largest Remainder

Overview

Allocate a quantized total by Hamilton largest remainder with deterministic keyed ties. Inputs are bounded plain decimals, output order follows input order, and no currency, policy, materiality threshold, market data, or ambient default is inferred.

For Allocate by Largest Remainder, we publish the exact bounded decimal rule and audit fields so you do not have to trust approximate prompt arithmetic.

When to use it

Scope

Non-goals

• Does not infer a rounding, materiality, currency, or business policy
• Does not fetch market data or invoke another public skill
• Does not accept unbounded arrays or binary floating-point semantics

Properties

Capabilities

Cost & billing

Flat rate per invocation

File handling

How we handle your data

Input schema

Output schema

API contract

curl -sS -X POST "$BASE/v1/execute" -H "Authorization: Bearer $KEY" -H 'Content-Type: application/json' -d '{"skillKey":"math.allocate_largest_remainder","input":{"total":"0.05","weights":[{"id":"B","weight":"1"},{"id":"A","weight":"1"},{"id":"C","weight":"1"}],"unitScale":2}}'

import { Squillify } from "@squillify/sdk";

const sq = new Squillify({ apiKey: process.env.SQUILLIFY_API_KEY });

const { output } = await sq.execute("math.allocate_largest_remainder", {"weights":[{"id":"B","weight":"1"},{"id":"A","weight":"1"},{"id":"C","weight":"1"}],"unitScale":2,"total":"0.05"});

Errors

Other limits

Limitations

• Does not infer a rounding, materiality, currency, or business policy
• Does not fetch market data or invoke another public skill
• Does not accept unbounded arrays or binary floating-point semantics
• When you want a continuous pro-rata split, or unitScale and unique ids are still undecided.
• Do not use for approximate floating-point analytics or unbounded datasets

How to call

POST /v1/execute (requires authentication)

{
  "skillKey": "math.allocate_largest_remainder",
  "input": {
    "weights": [
      {
        "id": "B",
        "weight": "1"
      },
      {
        "id": "A",
        "weight": "1"
      },
      {
        "id": "C",
        "weight": "1"
      }
    ],
    "unitScale": 2,
    "total": "0.05"
  }
}

Examples

{
  "weights": [
    {
      "id": "B",
      "weight": "1"
    },
    {
      "id": "A",
      "weight": "1"
    },
    {
      "id": "C",
      "weight": "1"
    }
  ],
  "unitScale": 2,
  "total": "0.05"
}

{
  "unit": "0.01",
  "ranking": [
    "A",
    "B",
    "C"
  ],
  "residualUnits": 2,
  "allocations": [
    {
      "id": "B",
      "amount": "0.02",
      "exactAmount": "0.01666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666667"
    },
    {
      "id": "A",
      "amount": "0.02",
      "exactAmount": "0.01666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666667"
    },
    {
      "id": "C",
      "amount": "0.01",
      "exactAmount": "0.01666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666666667"
    }
  ]
}

{
  "weights": [
    {
      "id": "north",
      "weight": "3"
    },
    {
      "id": "south",
      "weight": "1"
    }
  ],
  "unitScale": 0,
  "total": "10"
}

{
  "unit": "1",
  "ranking": [
    "north",
    "south"
  ],
  "residualUnits": 1,
  "allocations": [
    {
      "id": "north",
      "amount": "8",
      "exactAmount": "7.5"
    },
    {
      "id": "south",
      "amount": "2",
      "exactAmount": "2.5"
    }
  ]
}

Recommended use cases

• When a quantized total must be split into whole units with Hamilton remainders and keyed tie-breaks.
• Use in auditable fund-accounting calculations that need normalized decimal output

Avoid for

• When you want a continuous pro-rata split, or unitScale and unique ids are still undecided.
• Do not use for approximate floating-point analytics or unbounded datasets

Related skills

Composable with

Safety & privacy

• Runs without calling any external service.
• Your payloads are never stored permanently.
• Telemetry captures metadata only — never your payload contents.
• We never train models on your payloads.
• No human reviews your payloads.

Tags