unify

Infer and render conversational trajectories as diagrams, code, or type signatures.

1|Updated Mar 15, 2026

Search Tags:#diagrams #conversational-ai #pattern-recognition #unify #type-signatures #proof-trace #structure-rendering

Ready to run

npx skills add https://github.com/hafley66/claude-research --skill unify-hafley66

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: unify
Source: https://github.com/hafley66/claude-research/tree/main/skills/unify
Command: npx skills add https://github.com/hafley66/claude-research --skill unify-hafley66

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps users crystallize and render the underlying structure of a conversational trajectory, turning ideas into concrete, visible forms (diagrams, code, type signatures) to aid reasoning and decision-making.

Core Features & Use Cases

Idea unification: infer structure from an loosely described idea and render it as code, diagrams, or type signatures.
Conversation replay as proof search: replay turns as Prolog-style queries with bindings, show goals, bindings, open variables, and final open goal.
Flexible notation: adapt to domain (TS types, Prolog, architecture boxes).

Quick Start

Render the current conversational trajectory as a concrete structure (diagrams, code, or type signatures) to reveal the underlying pattern.

Frequently Asked Questions about unify

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I turn conversational ideas into concrete diagrams and type signatures?▼

To turn conversational ideas into concrete diagrams and type signatures, you can infer the underlying structure of a trajectory and render it as code blocks or diagrammatic boxes. This helps crystallize loosely described ideas into visible forms.

What's the best way to replay a conversation as a structured proof trace?▼

The best way to replay a conversation as a structured proof trace is to process the turns as Prolog-style queries with bindings. This method exposes goals, open variables, and the final open goal during rendering.

Can I render TypeScript type signatures from an unstructured conversational trajectory?▼

Yes, you can render TypeScript type signatures from an unstructured conversational trajectory. The system adapts to multi-domain notation, allowing you to infer structure and output it directly as TypeScript code blocks.

Does this approach support diagrammatic boxes for architecture rendering?▼

Yes, this approach supports diagrammatic boxes for architecture rendering. It applies flexible notation to adapt to your domain, outputting concrete representations as diagrams, code, or structured views.

When should I use pattern recognition to unify ideas in a conversation?▼

You should use pattern recognition to unify ideas when a user is circling an idea without landing on a concrete form. It helps infer the hidden structure and render it as visible diagrams or type signatures.