A systematic methodology for building reusable, agent-readable frameworks from any domain expertise.
Built by Mike Goetz. 600+ frameworks across industries since 2024.
A single SKILL.md file that teaches AI agents (and humans) how to build frameworks. It works in Claude Code, Claude Projects, Cursor, ChatGPT, Copilot, or any environment that supports skills or custom instructions.
The methodology produces five-layer frameworks that both agents and humans can follow for consistent results across any domain.
The AI ecosystem is converging on "skills" (markdown files that give AI context for specific tasks). Individual skills are useful, but they don't compound without systematic methodology.
This skill teaches the meta-methodology: how to generate frameworks across any domain. While everyone else is trading individual skills like baseball cards, this teaches you how to build the card factory.
Built from experience creating 600+ frameworks across federal contracting, sales methodology, medical advocacy, video production, content automation, and dozens of other industries.
- Drop
SKILL.mdinto your Claude Project, Cursor rules, or any AI skill directory - Tell your AI: "Build a framework for [your domain/problem]"
- The AI uses the five-layer DNA architecture to produce a complete, reusable framework
That's it. One file, any domain, consistent results.
Every framework generated by this methodology follows the same DNA:
- Layer 1: Principles Foundation - The WHY. Evidence-backed core truths, not opinions.
- Layer 2: Systematic Approach - The HOW. Reasoning steps with decision logic, not just procedures.
- Layer 3: Force Multipliers - The LEVERAGE. What creates outsized results from small inputs.
- Layer 4: Success Metrics - The MEASUREMENT. What good looks like, and what broken looks like.
- Layer 5: Implementation Guidance - The DEPLOYMENT. Edge cases, context adaptation, common mistakes.
See the examples/ folder for sample frameworks generated by this methodology.
Also see TRIGGER-MAP.md for when to reach for each framework in the companion meta-cognition series built by Grok (xAI).
The framework builder skill file teaches any AI how to generate a properly structured five-layer framework. It handles the architecture. The shape of what a framework should look like, what belongs in each layer, and how to make it agent-readable.
What it does not include is the intelligence library that makes SIOS-quality frameworks different from well-structured generic ones.
The Force Multiplier Catalog. A force multiplier is the mechanism that makes a framework compound beyond its basic application. Generic force multipliers like "automation," "delegation," and "leverage" are easy to generate. The SIOS catalog contains specific, cross-domain sourced force multipliers with exact mechanisms, documented failure modes, and known interaction effects with other frameworks. When an AI generates a framework without this catalog, it defaults to general examples. When it uses the catalog, the force multipliers are precise, non-obvious, and systematically chosen.
The Breakthrough Pattern Library. A library of documented patterns that identify when a situation is on the edge of a breakthrough. The signals, the conditions, the typical obstacles, and the moves that have worked historically across domains. Without this, frameworks can describe a process but cannot identify when the process is about to produce something exceptional.
The 343 Coordinate System. Every framework in SIOS receives four coordinates before layer construction begins: primary meta-category, cognitive foundations activated, transformation pattern, and contextual triggers. These coordinates shape the framework's design from the start rather than being appended afterward. Without the 343 architecture, frameworks are freestanding. With it, every framework knows where it sits in a connected system of 600+ others, and can activate the right neighbors automatically.
The Direction Transformer Catalog. A systematic library of transformation patterns. The specific moves that convert one type of problem into a different type of solvable problem. Cross-domain transfer, constraint inversion, scale shift, perspective rotation. Without this catalog, frameworks tend to address problems directly. With it, they can first transform the problem into a form where existing solutions already exist.
If you use the framework builder skill file to generate a five-layer framework, you will get a well-structured, agent-readable document with coherent principles, a systematic approach, and reasonable force multipliers. That is genuinely useful and significantly better than a prompt.
If you generate the same framework using the full SIOS methodology, with the catalogs, the coordinate system, and the pattern library, the force multipliers will be specific mechanisms drawn from adjacent domains, the coordinates will connect the framework to the broader library, and the breakthrough patterns will make the framework aware of its own edge conditions.
The open source release is the architecture. The methodology is what fills it.
Any AI using this skill file will produce frameworks based on its training data and whatever context you provide. The quality ceiling without the SIOS catalogs is the general knowledge of the model. The quality ceiling with the catalogs is the accumulated cross-domain intelligence of 600+ systematically built frameworks. Both are real. They are just different instruments.
This is the foundational framework generation process. For the full ecosystem:
- Advanced multi-framework orchestration: ragedesigner.com — Strategic consulting and the complete SIOS framework library (600+ frameworks)
- Learn the methodology: howtoframework.com — Practical guides for building frameworks from your own domain expertise
- Understand what frameworks are: whatisaframework.com — The foundational concepts behind systematic thinking
- Find frameworks by use case: whereframeworks.com — Browse frameworks organized by industry and problem type
- Courses and learning paths: whenframework.com — Structured courses including Love Your Limits (constraint-driven innovation) and Reverse Engineering Success
MIT License. Use it, share it, build on it.
Mike Goetz is the founder of RageDesigner (est. 2003) and creator of the Strategic Intelligence Operating System (SIOS). He has built 600+ frameworks across unlimited industries using this methodology since 2024. His work bridges the gap between academic research and practical business application through systematic framework generation.