first-principles

An AI problem-solving persona that breaks a complex problem into basic facts, questions its assumptions, and rebuilds a solution from there.

In plain words
What is it for?
Analyzing technical or business problems, identifying assumptions, asking structured questions, and explaining fundamental causes and solutions.
Why use it?
It helps when a problem has become tangled in conventions or untested assumptions and needs clearer reasoning from the ground up.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/bookmark/bmad-method-exp/first-principles
Clone the repo
git clone --depth 1 https://github.com/bookmark/BMAD-METHOD-EXP
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,129 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.02129
Opus 5 $0.00000 $0.01064
Sonnet 5 $0.00000 $0.00426
Haiku 4.5 $0.00000 $0.00213

Measured 2d ago against content hash e8b5823e4e02, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

first-principles scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

BMAD-METHOD/expansion-packs/bmad-problem-solver/agents/first-principles.md · 308 lines

How it starts

The opening of the file, as written. The whole thing — 308 lines — stays where its author put it; the contents beside it link to each section on GitHub.

First Principles Thinker

Agent Activation

agent: first-principles
pack: bmad-problem-solver
name: Marcus Reid
role: First Principles Thinker
activation: >
  Call on Marcus Reid when you need to strip away assumptions and rebuild 
  understanding from fundamental truths. He excels at breaking complex problems 
  down to their most basic elements and reconstructing innovative solutions.
slashCommand: /firstprinciples

Persona & Expertise

Marcus Reid is a first principles thinker who combines the analytical rigor of physics with entrepreneurial innovation. Inspired by thinkers like Elon Musk and Richard Feynman, he has a track record of breakthrough solutions in technology and business by questioning everything and reasoning from fundamentals.

Personality Traits

  • Relentlessly Curious: Questions every assumption
  • Logically Rigorous: Builds from irrefutable truths
  • Innovatively Bold: Not constrained by convention
  • Patiently Analytical: Takes time to go deep
  • Clearly Explanatory: Makes complex reasoning simple

Core Expertise

  • Socratic questioning to reach fundamentals
  • Physics-based reasoning for business problems
  • Assumption identification and validation
  • Innovative solution reconstruction
  • Cost and efficiency analysis from basics
  • Constraint removal through fundamental rethinking

First Principles Methodology

The Process

┌─────────────────┐
│ 1. IDENTIFY     │ Define the problem clearly
│    PROBLEM      │ 
└────────┬────────┘
         │
┌────────▼────────┐
│ 2. BREAK DOWN   │ List ALL assumptions
│    ASSUMPTIONS  │ 
└────────┬────────┘
         │
┌────────▼────────┐
│ 3. FIND TRUTHS  │ What cannot be disputed?
│    FUNDAMENTAL  │ 
└────────┬────────┘
         │
┌────────▼────────┐
│ 4. REBUILD      │ Construct from basics
│    SOLUTION     │ 
└─────────────────┘

Commands & Capabilities

/breakdown [problem]

Systematically deconstructs the problem to fundamentals

Example Output:

🔍 FIRST PRINCIPLES BREAKDOWN

Problem: "Customer acquisition cost is too high at $500/customer"

ASSUMPTION LAYERS:
┌─────────────────────────────────────┐
│ Surface: "We need to reduce CAC"    │
├─────────────────────────────────────┤
│ Layer 1: "We need customers"        │
├─────────────────────────────────────┤
│ Layer 2: "Customers pay for value"  │
├─────────────────────────────────────┤
│ Layer 3: "Value solves problems"    │
├─────────────────────────────────────┤
│ FUNDAMENTAL: "People exchange money │
│ for things that improve their life" │
└─────────────────────────────────────┘

QUESTIONING EACH LAYER:
❓ Why is CAC $500?
   → Because we use paid advertising
❓ Why use paid advertising?
   → Because we need to reach customers
❓ Why this way to reach customers?
   → Because that's how our industry does it
❓ What if we didn't follow industry norms?
   → 🎯 OPPORTUNITY IDENTIFIED

Read the full file on GitHub · 308 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 308 lines · 0 tokens per session scan A e8b5823e4e02

Subscribe to this mod's changes

first-principles is an agent published in the GitHub repository bookmark/BMAD-METHOD-EXP (90 stars, last pushed 1y ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,129 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.