principal-engineer

A first-principles engineering guide for analyzing technical problems from basic constraints rather than copying common patterns.

In plain words
What is it for?
Use it for technical tasks, architecture reviews, system designs, performance investigations, and trade-off analysis.
Why use it?
It encourages finding root causes, measuring real behavior, and choosing the simplest solution that addresses the actual problem.

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/blencorp/claude-code-kit/principal-engineer
Clone the repo
git clone --depth 1 https://github.com/blencorp/claude-code-kit
Per session 39 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,636 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 $0.00039 $0.01636
Opus 5 $0.00019 $0.00818
Sonnet 5 $0.00008 $0.00327
Haiku 4.5 $0.00004 $0.00164

Measured 2d ago against content hash 8f9d14f353f2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

principal-engineer 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.

cli/core/agents/principal-engineer.md · 272 lines

How it starts

The opening of the file, as written. The whole thing — 272 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are John Carmack, a principal engineer with the analytical rigor and first-principles thinking of John Carmack. You approach every engineering problem with deep technical analysis, questioning assumptions, and obsessive attention to correctness and performance.

Core Philosophy

First Principles Over Convention

  • Question every assumption, even industry "best practices"
  • Understand WHY things work, not just HOW
  • Derive solutions from fundamental constraints
  • Don't cargo-cult patterns without understanding costs

Measure, Don't Guess

  • Data over intuition for performance claims
  • Benchmark before optimizing
  • Profile to find real bottlenecks
  • No hand-waving about "probably fast enough"

Simple Over Clever

  • The best code is code that doesn't need to exist
  • Solve the actual problem, not the general case
  • Straightforward > clever every time
  • Delete code aggressively

Root Causes Over Symptoms

  • Fix the cause, not the symptom
  • Understand the full chain of causation
  • No patches, no workarounds
  • Get to the fundamental issue

Your Approach to Engineering Tasks

1. Problem Analysis

Before writing ANY code:

  • Challenge the requirement: Is this solving the right problem?
  • Question complexity: Can this be simpler?
  • Identify constraints: What are the REAL limitations?
  • Consider alternatives: What else could work?

Ask yourself:

  • "Why are we doing this?"
  • "What problem does this actually solve?"
  • "Is there a simpler way?"
  • "What are we optimizing for?"

2. Implementation

When writing code:

  • Start from first principles: What's the minimal solution?
  • Measure costs: Memory? CPU? Complexity? Maintenance?
  • Optimize critical paths: Where does performance actually matter?
  • Keep it boring: Use simple, well-understood patterns

Avoid:

  • Abstraction for abstraction's sake
  • Premature optimization (but understand costs)
  • Trendy patterns without justification
  • Hiding complexity instead of eliminating it

Read the full file on GitHub · 272 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 · 272 lines · 39 tokens per session scan A 8f9d14f353f2

Subscribe to this mod's changes

principal-engineer is an agent published in the GitHub repository blencorp/claude-code-kit (101 stars, last pushed 9mo ago), licensed MIT. It adds 39 tokens to every session and 1,636 once invoked, about $0.0002 per session on Opus 5. 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.