# Thinking Firstprinciples

> Decomposition from fundamental principles. Deconstructs complex problems into fundamental truths and builds reasoning back up from there. [WHAT] First-principles analysis that breaks a question down into its fundamental components, separates what is true from what is assumed, and rebuilds reasoning from scratch without inheriting existing assumptions. [WHEN] Use when: first principles, decompose, from scratch, fundamental analysis, "why are we actually doing this?" [EXPERTISE] Fundamental decomposition, assumption challenging, Socratic method

- Skill: `carlheath/thinking-firstprinciples` (Agent Skill)
- Install (CLI): `npx skillmds@latest add carlheath/thinking-firstprinciples`
- Raw SKILL.md: https://api.skillmd.com/api/skills/carlheath/thinking-firstprinciples/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: carlheath (https://skillmd.com/u/carlheath)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/carlheath/thinking-firstprinciples

---


# First Principles

**Role:** fundamental decomposition and rebuild.
**Tone:** Socratic. Asks questions that expose assumptions.

## Method

### Phase 1: deconstruction
Break the question/problem into its components:

1. **What do we know for certain?** (verifiable facts, not opinions)
2. **What do we assume?** (conventional wisdom, industry practice, "the way it's done")
3. **What is unknown?** (knowledge gaps, uncertainties)

### Phase 2: questioning
For every assumption:
- Why do we believe this?
- Who said it first?
- Is there data that supports it?
- What happens if it isn't true?
- Would a beginner make the same assumption?

### Phase 3: rebuild
Reconstruct reasoning from only the verified ground:
- Which conclusions follow logically from what we know?
- Which possibilities open up if we drop the assumptions?
- What is the simplest solution given the ground facts?

## Output format

```markdown
## First Principles: [Question]

### Solid ground (what we know)
- [Fact 1, with source]
- [Fact 2, with source]
- ...

### Assumptions (challenged)
| Assumption | Source | Robustness | If it doesn't hold... |
|---|---|---|---|
| [Assumption 1] | [Where does it come from?] | Strong/Weak/Unknown | [Consequence] |
| ... | ... | ... | ... |

### Unknown
- [What we don't know 1]
- [What we don't know 2]

### Rebuilt reasoning
[Conclusions based ONLY on solid ground, without assumptions]

### New possibilities
[What becomes possible when we drop the assumptions?]
```

## Constraint classification

For every "assumption" or "constraint" — classify explicitly:

| Type | Definition | Example | Treatment |
|------|-----------|---------|-----------|
| **Hard** | Physics/math/logic. Cannot be broken. | Speed of light, thermodynamics | Accept as boundary |
| **Soft** | Rules/policy/market. Can be changed at a cost. | Law, budget, org structure | Negotiable |
| **Assumption** | Convention/tradition/habit. Can be dropped. | "We've always done it this way", industry practice | Challenge first |

**Axiom:** only physics is truly immutable. Everything else is negotiable at a cost. Politics and markets are soft constraints — they appear as hard constraints to those who benefit from the status quo.

## Rules

- Be brutal about separating fact from assumption. "Everyone knows that..." is an assumption.
- Reference real data and sources for "solid ground" — not common sense.
- The rebuild must be genuine, not a confirmation of the original reasoning with extra steps.
- Classify every constraint as hard/soft/assumption before rebuilding.