What is the derangement constraint in this skill?

The derangement constraint (σ(i)≠i) ensures that information never self-loops; every bit of data must flow to a different position or agent to prevent stasis and bias.

How does tropical geometry apply to Claude Code skills?

It uses min-plus semiring math to calculate the 'shortest path' for task completion while ensuring no consecutive tasks share the same role, minimizing interaction costs.

How does GF(3) accounting improve workflows?

It categorizes actors into three roles: Validators (-1), Coordinators (0), and Generators (+1). By ensuring the sum of these roles is conserved, the system maintains a balanced flow of information.

What is a 'fixed-point pathology' in code reviews?

It occurs when a validator checks their own logic or patterns, leading to a lack of objective oversight. Derangement Reflow prevents this by routing output to distinct validators.

Derangement Reflow Optimizer

Name: Derangement Reflow Optimizer
Author: plurigrid

byplurigrid

•

Productivity & Workflow

Optimizes multi-agent information flow by enforcing derangement constraints to prevent self-validation loops and ensure dynamic entropy redistribution.

The Derangement Reflow skill provides a mathematical framework for managing information exchange within complex workflows. By leveraging GF(3) triadic accounting and tropical geometry, it identifies and fixes 'fixed-point pathologies'—situations where agents or validators unintentionally self-validate their own patterns. This skill is essential for developers building robust multi-agent systems, as it ensures that information always flows to a different position in the graph (σ(i)≠i), maximizing joint world model accuracy and minimizing surprise through Active Inference principles.

Key Features

01Joint world modeling via Active Inference and free energy minimization

02Fixed-point pathology detection to identify biased validation loops

03Tropical geometry path optimization for minimal interaction cost

04Automated derangement validation for multi-skill triplet compositions

052 GitHub stars

06GF(3) trit-based task accounting (Minus, Ergodic, Plus)

Use Cases

01Eliminating self-validation bias in multi-agent software testing and code review pipelines

02Designing high-entropy information architectures for complex system modeling

03Optimizing task interleaving between generators, coordinators, and validators

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add plurigrid/asi derangement-reflow

For use in Claude.ai and ChatGPT

Download Skill