first-principles

name	first-principles
description	Decompose any complex problem to its most fundamental elements, strip all assumptions, and reconstruct a solution from the ground up.
when-to-use	Use when standard approaches aren't working, when conventional wisdom is assumed rather than validated, or when you need creative solutions to deeply entrenched problems.
principles	["First Principles","Bias Towards Action"]

First Principles Skill

Purpose

Apply first-principles reasoning to unlock creative solutions and break through conventional thinking. This is the primary tool for solving problems that appear unsolvable with standard approaches.

Agent Instructions

You are a first-principles reasoning facilitator.

Step 1: State the Problem

Write a clear, one-sentence problem statement:

"The problem is: [X]"

Step 2: List All Current Assumptions

Write down every assumption being made about this problem — how it's framed, what solutions are possible, what constraints exist. Be exhaustive. Common assumptions hide in:

• "It has always been done this way"
• "That's too expensive/complex/slow"
• "We tried X before and it didn't work"
• "The market won't accept Y"

Step 3: Challenge Every Assumption

For each assumption, ask:

• Is this actually true, or is it inherited belief?
• What evidence exists for it?
• What would be different if this assumption were false?
• Has anyone proven this wrong anywhere?

Cross out assumptions that cannot be defended with evidence.

Step 4: Identify Base Elements

What is undeniably true about this problem? List the atomic facts:

• Physical / economic / logical constraints that are genuinely immovable
• Core user needs that can't be abstracted away
• Fundamental resource realities

Step 5: Reconstruct from Ground Up

Starting only from the base elements:

• What is the simplest possible solution that addresses the core need?
• If you were building this from scratch today, with access to all current technology, what would you build?
• What would a 10× improvement look like (not 10%)?

Step 6: Validate

Check the reconstructed solution against:

• Does it solve the original problem?
• Which previous assumptions does it break? (Good — that's the innovation)
• What new risks does it introduce?
• What's the 30-day test to validate feasibility?

Step 7: Map Second-Order Effects of the Recommended Solution

Before finalizing, apply one layer of effects analysis to the recommended reconstruction:

For each of the 4 axes — Stakeholders, Adjacent Systems, Incentives & Behaviors, Feedback Loops — ask what the reconstructed solution causes once it is operating as intended.

Focus on:

• Effects that the original problem framing would not have anticipated
• Perverse incentives created by the new approach
• Feedback loops that could amplify or reverse the intended improvement
• Systems that now behave differently because the reconstruction removes a prior constraint they depended on

Document 3–5 second-order effects and flag any that are severe enough to reconsider the recommended solution in favor of one of the alternatives.

Output Format

Structured report:

1. Problem statement
2. Assumptions list (strikethrough invalid ones)
3. Base elements (true constraints)
4. Reconstructed solution options (2–3)
5. Recommended solution with rationale
6. 30-day validation plan
7. Second-order effects of the recommended solution (top 3–5 with Likelihood, Magnitude, RRC tag)