Skip to content

Latest commit

 

History

History

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 

README.md

Chapter 02 — The 5-prompt frame

This chapter is the methodology core of the whole repo. Everything else builds on the structure described here. Worth reading slowly — or twice, if you came here from chapter 1 already convinced.

Files

  • frame.md — the full essay: why these five dimensions, how to calibrate findings, what makes a finding good vs bad, and a 10-minute rubric for triaging
  • The prompt itself sits at ../prompts/system-prompt.md (CC0)

What you walk away with

  1. Why these 5 dimensions catch what OWASP / SRE checklists miss
  2. How to recognise a bad finding ("add monitoring" with no thresholds) and reject it
  3. A rubric for triaging findings before they hit severity ranking

TL;DR

The frame asks, for any design, AT LEAST 2 concrete failure scenarios across each of these five dimensions:

  1. Hidden assumptions — ordering, uniqueness, atomicity, data freshness, caller behavior
  2. Dependency failures — what happens when upstream / downstream services degrade or go down
  3. Boundary inputs — empty, single, huge batch, malicious, malformed
  4. Misuse paths — caller / user / operator using it the wrong way
  5. Rollback & blast radius — how do you recover, and how big is the damage if it goes wrong?

Every scenario must include TRIGGER, IMPACT, and DETECTABILITY. Generic advice like "add monitoring" is rejected outright — if you can't say what metric, what threshold, what alert, the finding isn't ready.

frame.md has the full reasoning behind each dimension, calibration examples, and the 10-minute rubric.

← Back to README · 中文版