How does this skill support refinement types?

It automatically derives more specific types based on proven constraints, such as refining a standard Integer type into an 'EvenPositiveInt' after it passes through specific logic filters.

What is constraint generalization in the context of Claude Code?

It is a skill that allows Claude to synthesize new logical constraints that hold true when multiple code paths or functions are combined, enabling deeper static analysis and automated reasoning.

Is this skill useful for performance optimization?

Yes. By proving constraints once and propagating them downstream, it identifies redundant checks that can be safely removed and enables smarter composition of navigation paths.

What happens if two code paths have contradictory constraints?

The skill utilizes an underlying 3-MATCH SAT solver to detect unsatisfiable compositions, identifying logical errors (like a value needing to be both even and odd) during the design phase.

Constraint Generalization Logic

Name: Constraint Generalization Logic
Author: plurigrid

byplurigrid

•

開発者ツール

Synthesizes and propagates formal constraints across complex code paths to enable automatic refinement typing and proof generation.

Constraint Generalization is a sophisticated logic skill that enhances Claude's ability to handle complex data transformation paths by moving beyond simple validation to active constraint synthesis. Drawing from constraint logic programming and abstract interpretation, it allows Claude to derive new, higher-level properties when multiple code paths are composed—such as automatically identifying that a value must be both even and positive. This skill enables the generation of refinement types and formal proof certificates, utilizing a 3-MATCH SAT solver to ensure logical consistency and eliminate redundant runtime checks in high-integrity software systems.

主な機能

01CNF-based composition to detect and prevent logical contradictions

02Abstract interpretation for constraint weakening and lattice-based generalization

03Formal proof certificate generation using SAT-based logic

04Automatic constraint synthesis across composed navigation paths

057 GitHub stars

06Refinement type generation for enhanced static analysis

ユースケース

01Formal verification of complex business logic and backend flows

02Building type-safe data pipelines with automatic property propagation

03Automating edge-case discovery via constraint satisfiability checks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add plurigrid/asi constraint-generalization

For use in Claude.ai and ChatGPT

Download Skill