first-principles-thinking · diff
v2026.5.17 to v2026.5.17
8 added, 4 removed. Audit A to A.
---
name: first-principles-thinking
description: >
- Strip analogies and convention to surface fundamental truths, then rebuild a
- reasoning chain and implications. Use when the user asks for first principles,
- "reason from scratch", challenge industry defaults, physics-of-business style
- breakdowns, or to sanity-check whether copying incumbents makes sense.
+ Use this skill when the user asks for first-principles thinking or wants to
+ reason from bedrock—stripping borrowed analogies and convention, surfacing
+ fundamentals, then rebuilding the reasoning chain and implications. Use when
+ they want to reason from scratch, challenge industry defaults, want physics-style
+ business breakdowns, or sanity-check whether copying incumbents still makes
+ sense, even if they never say first principles. Skip when they want a quick
+ convention-following checklist with no rebuild of assumptions, or purely
+ social coordination with no modeling ask.
license: MIT
metadata:
author: ysskrishna
version: "2026.5.17"
---
# First Principles Thinking
Question inherited baggage. Rebuild only from **bedrock** you can defend.
---
## When to Use
Use when analogies may mislead: new markets, hard tech, cost structure invention, personal strategy against social scripts. Less useful when best practice is truly optimal and context-free.
---
## Setup (run before starting)
In one short block:
1. **Reconstruction target** — belief, cost, design, or strategy to ground
2. **Default pass** — Surface → Question → Bedrock → Rebuild → Implications (state this line)
Up to 3 questions on immutable constraints (physics, law, budget); mark unknowns `[UNKNOWN]`.
---
## The Steps
### Surface
State the **conventional answer** or analogy people rely on. List **loaded words** or hidden comparisons ("like Uber for…").
### Question
For each major assumption: **Assumption:** … — **Why believed?** (authority, analogy, experience) — **What if false?**
### Bedrock
List **fundamental truths** that survive scrutiny — physics, logic identities, legal musts, documented preferences of real users, arithmetic. Label each `[FUNDAMENTAL]` vs `[STILL ASSUMPTION]`.
Aim for **at least 3** bedrock items; if fewer are honest, explain why.
### Rebuild
From **only** `[FUNDAMENTAL]` items, derive conclusions in numbered steps. No smuggled analogies; if you need a new premise, add it to **Bedrock** first with a tag.
### Implications
**So what** for decisions: what changes vs the conventional path? **Cost of being wrong** if a tagged assumption fails.
Add a short **vs convention** contrast (a few bullets or a two-column mini-summary) if it clarifies the decision.
---
## Execution Rules
1. **Rebuild** cannot cite "industry standard" as a premise unless translated into a fundamental (e.g. "buyers require SLA X because regulation Y").
2. If bedrock is too thin to rebuild, say **insufficient grounding** and list what evidence would fix it.
3. Avoid faux profundity; keep steps short and checkable.
---
## Checklist (verify before responding)
- [ ] Setup: reconstruction target + default pass
- [ ] Surface names convention or analogy explicitly
- [ ] Question step ties assumptions to why held
- [ ] Bedrock uses `[FUNDAMENTAL]` / `[STILL ASSUMPTION]`
- [ ] Rebuild chain only uses fundamentals
- [ ] Implications name what changes vs convention (and vs-contrast if helpful)