Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Fair Split

ASecurity

Divide things that cannot be divided — an inherited house, a dissolved partnership, the furniture after a breakup — using sealed-bid fair division (Knaster), computed by the bundled script so everyone ends provably above their own fair share. Use when asked to split an estate fairly, divide assets between siblings or ex-partners, who gets the house and what do they owe the others, or handle unequal shares. Produces the sealed-bid procedure, the computed allocation with exact cash transfers, t...

1,378 stars
0 votes
0 copies
0 views
Added 9/20/2026
ai-agentspythongoreactgit

Security Analysis

A100/100

Scanned 9/20/2026

$npx -y skills add mohitagw15856/pm-claude-skills --skill fair-split --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Fair Split?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Fair Split
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/mohitagw15856-fair-split/badge)](https://www.skillsdirectory.com/skills/mohitagw15856-fair-split)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: fair-split
description: "Divide things that cannot be divided — an inherited house, a dissolved partnership, the furniture after a breakup — using sealed-bid fair division (Knaster), computed by the bundled script so everyone ends provably above their own fair share. Use when asked to split an estate fairly, divide assets between siblings or ex-partners, who gets the house and what do they owe the others, or handle unequal shares. Produces the sealed-bid procedure, the computed allocation with exact cash transfers, the per-person fairness proof, and the presentation script. Math for the family meeting; not legal advice."
homepage: https://mohitagw15856.github.io/pm-claude-skills/skill/fair-split.html
metadata:
  {
    "openclaw": { "emoji": "🧮" }
  }
---

# Fair Split

The house cannot be cut in three, everyone valued it differently, and the sibling who wants it least still resents the one who got it — that is how estates turn into estrangements. Sealed-bid fair division (the Knaster procedure) fixes the structural problem: each person privately states what every item is worth *to them*, each item goes to whoever values it most, and cash transfers rebalance so that every single person ends up — by their own numbers — with more than their fair share. The person who wanted nothing gets paid; the person who got the house pays for the privilege at their own valuation. The bundled script does the arithmetic exactly, supports unequal entitlements (a 50/25/25 will, a 60/40 partnership), and proves the outcome to each participant in their own terms.

## What This Skill Produces

- **The procedure to run** — how to collect sealed valuations so they are honest: privately, simultaneously, before anyone knows the mechanism rewards sincerity
- **The computed allocation** — who gets each item and the exact cash each person pays or receives, from the script
- **The proof of fairness, per person** — each participant shown ending above *their own* valuation of their fair share, which is the sentence that ends arguments
- **The presentation script** — how to explain the result to people who are grieving, angry, or both
- **The edge-case handling** — the item someone cannot afford to win, the sentimental object, the participant who refuses to play

## Required Inputs

Ask for these if not provided:
- **The people and their shares** — who is dividing, and the entitlements (equal by default; a will or partnership agreement may set them unequal)
- **The items** — everything indivisible worth listing; small stuff can be bundled into lots
- **The valuations** — each person's private value for every item; if not yet collected, the skill's first output is the collection procedure, because *how* they are collected determines whether they are honest
- **The constraints** — anyone cash-poor (winning the house means paying the others — can they?), any item with a legal restriction, any participant acting for someone else

## Framework: Sealed Bids, Highest Valuer Wins, Cash Rebalances

1. **Collect valuations sealed and simultaneous.** The mechanism only rewards honesty if nobody bids in reaction to anyone else. Each person answers privately: *what is this worth to you, in money, knowing you might have to pay that*. The framing matters — it is not a guess at market price, it is their personal price.
2. **Run the script — the arithmetic is not negotiable, which is its gift.**
   ```
   python3 scripts/fair_split.py --input split.json          # table
   python3 scripts/fair_split.py --input split.json --json   # for further processing
   python3 scripts/fair_split.py --demo                      # worked three-way estate example
   ```
   Each item goes to its highest valuer; each person's fair share is their entitlement weight times their own total valuation; the surplus the assignment creates is shared back by weight; cash transfers settle the difference and provably sum to zero.
3. **Present each person their own proof.** "By the numbers *you* gave, your fair share was X — you ended with X plus Y." Fairness argued from the other side's numbers convinces nobody; fairness proven from their own numbers has nothing left to argue with.
4. **Handle the cash-poor winner before running it.** Winning the big asset means paying real money to the others. If that is impossible, options: sell the asset and split proceeds (list it at everyone's-bid as reserve), a payment plan the others accept, or exclude it from the procedure and handle it separately.
5. **Respect the sentimental override.** The mechanism prices sentiment honestly — the person who bids high on the photo albums *pays* for them, which is usually right. But an item everyone agrees belongs to one person should be gifted before the procedure, not run through it.
6. **The refuser.** Someone who won't participate can be represented by market valuations with their consent, or the procedure runs on the subset who will — but a division imposed on a non-participant is not fair division, and the skill says so.

## Output Format

### Fair split: [what is being divided] · [n] people · [date]

**Entitlements:** [name: %] · **Items:** [count] · **Valuations collected:** [sealed/simultaneous — or the collection procedure to run first]

**The allocation** (from `fair_split.py`)
| Person | Share | Items won | Cash (+receives/−pays) | Ends with (own valuation) | Above own fair share by |
|---|---|---|---|---|---|

**The proof, per person:** [one sentence each, in their own numbers]

**Flags:** [cash-poor winner and the options · sentimental pre-gifts made · non-participants and how handled]

**Presenting it:** [the script — lead with the guarantee, show each their own proof, the arithmetic-is-neutral framing]

> Math for the family meeting, not legal advice. Estates in probate, tax consequences of transfers, and jointly-owned property have legal machinery this procedure does not replace — run the numbers, then run the result past the professional handling the estate.

## Quality Checks
- [ ] Valuations were (or will be) collected sealed and simultaneous, and the output says why
- [ ] The script's transfers sum to zero and every person ends above their own fair share
- [ ] Unequal entitlements, if any, are reflected in the weights, not bolted on after
- [ ] The cash-poor-winner check ran before the result was presented
- [ ] Each person's proof uses their own valuations only
- [ ] Legal-machinery boundaries are flagged

## Anti-Patterns
- **Collecting bids in the open.** One overheard number corrupts every later one, and the honesty guarantee dies.
- **Explaining fairness with the aggregate.** Each person cares about their own numbers; prove it person by person.
- **Running the un-runnable item.** If nobody can pay for the house, the procedure just formalises a crisis — handle liquidity first.
- **Pricing what should be gifted.** The mechanism is for contested value, not for making a sister buy her own childhood letters.
- **Presenting the output as a verdict.** It is a proposal with unusually good properties; adults still get to say no.
- **Using it to steamroll a grieving refuser.** Consent to the mechanism is part of the mechanism.

Attribution

mohitagw15856mohitagw15856
View sourceSee grades on GitHubMore from mohitagw15856 →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698461 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →