Twenty minutes and want the heart of the argument
ThenChapter 5 and the take-home kit (Ch. 11)
§ 01 — Start here
Where to start
Pick the row that sounds like you. It tells you which chapter to read first, and where to go after.
01 · You have
02 · Start with
03 · Leads to
Open
An agent about to get production credentials
Agents burning tokens on context I already have in source
ThenLean AGENTS.md; add layers only when they pay — see the take-home kit
Changes that depend on a repository the agent cannot read
ThenContracts and freshness gates in Chapter 5
No clear “done” before the agent starts writing
Reports that say passed while the real check is unverified
ThenIndependent checks the agent cannot rewrite — stay in Chapter 5
The same environment failures every agent session
ThenFixes in the harness, not only in prompts
Need to match review depth to blast radius
ThenTiered review from reversibility and verification strength — Chapter 8
Wondering whether a set-up change actually helped
ThenRepeated runs with stated limits — Chapter 9
Ready to adopt — what do we put in place first
ThenDefault / Conditional / Experimental labels for each layer — Chapter 11
Want the findings and thesis before any chapter
ThenEleven chapters with checklists — from Chapter 1 onward, or the Guide contents
Numbers in chips (for example +35–48%,14 / 16, 3 → 18–19/20,12 runs) come from the Foreword findings. Each has limits stated in the Guide.