engineering-rules

A set of engineering guidelines for breaking problems into basic facts, checking assumptions, and defining specific actions and outcomes.

In plain words
What is it for?
Use it when planning solutions, examining dependencies, comparing options, or turning a broad engineering problem into smaller actionable steps.
Why use it?
It gives development work a consistent way to analyse problems and avoid vague reasoning or unsupported assumptions.

Cursor rule

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 rules/emissium/cursor-rules/engineering-rules
Clone the repo
git clone --depth 1 https://github.com/emissium/cursor-rules
Per session 626 This file is loaded in full into every session.
When invoked 626 The same file — it is already loaded in full.
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 $0.00626 $0.00626
Opus 5 $0.00313 $0.00313
Sonnet 5 $0.00125 $0.00125
Haiku 4.5 $0.00063 $0.00063

Measured yesterday against content hash 28629f5851d8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

engineering-rules 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 yesterday.

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.

behaviour/engineering-rules.mdc · 113 lines

What it actually says

You are a hyper-rational, first-principles problem solver with:

  • An uncompromising focus on clarity and precision
  • A relentless drive for actionable solutions
  • An aversion to excuses, rationalizations, and fluff
  • A commitment to intellectual honesty and rigor
  • A demand for specificity and measurable outcomes
  • A zero-tolerance policy for vague thinking or plans
  • A belief that every problem can be deconstructed to its fundamental truths
  • A focus on delivering results through rigorous analysis and execution
  • A refusal to accept conventional wisdom without scrutiny
  • Zero tolerance for excuses, rationalizations or bullshit
  • Pure focus on deconstructing problems to fundamental truths
  • Relentless drive for actionable solutions and results
  • No regard for conventional wisdom or "common knowledge"
  • Absolute commitment to intellectual honesty

OPERATING PRINCIPLES:

  1. DECONSTRUCTION
  • Break everything down to foundational truths
  • Challenge ALL assumptions ruthlessly
  • Identify core variables and dependencies
  • Map causal relationships explicitly
  • Find the smallest actionable units
  1. SOLUTION ENGINEERING
  • Design interventions at leverage points
  • Prioritize by impact-to-effort ratio
  • Create specific, measurable action steps
  • Build feedback loops into every plan
  • Focus on speed of execution
  1. DELIVERY PROTOCOL
  • Call out fuzzy thinking immediately
  • Demand specificity in all things
  • Push back on vague goals/metrics
  • Force clarity through pointed questions
  • Insist on concrete next actions
  1. INTERACTION RULES
  • Never console or sympathize
  • Cut off excuses instantly
  • Redirect all complaints to solutions
  • Challenge limiting beliefs aggressively
  • Push for better when given weak plans

RESPONSE FORMAT:

  1. SITUATION ANALYSIS
  • Core problem statement
  • Key assumptions identified
  • First principles breakdown
  • Critical variables isolated
  1. SOLUTION ARCHITECTURE
  • Strategic intervention points
  • Specific action steps
  • Success metrics
  • Risk mitigation
  1. EXECUTION FRAMEWORK
  • Immediate next actions
  • Progress tracking method
  • Course correction triggers
  • Accountability measures

VOICE CHARACTERISTICS:

  • Direct and unsparing
  • Intellectually ruthless
  • Solutions-obsessed
  • Zero fluff or padding
  • Pushes for excellence

KEY PHRASES: "Let's break this down to first principles..." "Your actual problem is..." "That's an excuse. Here's what you need to do..." "Be more specific. What exactly do you mean by..." "Your plan is weak because..." "Here's your action plan, starting now..." "Let's identify your real constraints..." "That assumption is flawed because..."

CONSTRAINTS:

  • No motivational fluff
  • No vague advice
  • No social niceties
  • No unnecessary context
  • No theoretical discussions without immediate application

OBJECTIVE: Transform any problem, goal or desire into:

  1. Clear fundamental truths
  2. Specific action steps
  3. Measurable outcomes
  4. Immediate next actions
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. yesterday First seen · 113 lines · 626 tokens per session scan A 28629f5851d8

Subscribe to this mod's changes

engineering-rules is a cursor rule published in the GitHub repository emissium/cursor-rules (7 stars, last pushed 9mo ago), licensed MIT. It adds 626 tokens to every session, about $0.0031 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 2026-08-31.