Nothing here is a summary written after the fact. These are the problem, mechanism,
and move of each article, in the reading order, joined by its own connective lines.
Any one of them opens into the full argument.
Start with what actually rots, because the answer decides everything downstream: not judgment anyone lacked, but information nobody could hold.
Good engineers, thorough reviews, and competent organizations still produce systems that degrade, and the moral explanations, carelessness and inexperience, do not survive contact with the actual cases.
Mechanism
The information that would have prevented the failure exists but sits outside the head making the decision, and no amount of experience adds working-memory slots to hold it, because expertise enlarges what fits in a slot rather than adding one.
Move
Manufacture the presence of context at the point of decision instead of assuming it, and codify a pattern before asking any tool to enforce it.
If the problem is holding information rather than producing work, then the half of the job that just became free was never the scarce half.
Every methodology in wide use was designed around an assumption that has quietly stopped being true, that producing the artifact is the expensive part of producing the artifact.
Mechanism
When one input to a process falls to nearly nothing and the others hold, the process reorganizes around what is still scarce, and what is still scarce is establishing that the thing is sound, because the cost of a defect that escapes did not fall with the cost of making one.
Move
Spend the freed hours on risk analysis rather than on more output, and treat the review of generated work as the primary contribution rather than the tax on it.
Verification is now the job, which raises the harder question of why it is so difficult to perform on work that looks this good.
Generated work arrives carrying every signal a decade of practice taught practitioners to trust, and those signals were all calibrated against human error, which looks nothing like this.
Mechanism
A false premise adopted early is built on consistently, so the artifact ends up internally coherent with its own error, and each conventional check confirms that coherence instead of catching the premise.
Move
Verify the premises rather than the behaviour, read the tests before the implementation, and ask a separate session to explain the work back rather than to approve it.
The illusion is one decay and there is a second, slower one that runs on the clock and produces code no reviewer will stop.
Practitioners guard long working sessions against the model getting things wrong, and the thing that actually decays across a long session is quieter than error and slides straight past review.
Mechanism
As a session extends and its context fills, the specific constraints that make a codebase yours compete with the far larger mass of convention the model was trained on, and convention wins by weight.
Move
Keep sessions atomic and start fresh ones more often than feels necessary, feed only the subsystem a task actually needs, and hold the standing constraints in a file that is reloaded rather than in a conversation that is remembered.
End where all of it lands if the checking never happens, with a system that runs perfectly and that nobody has ever understood.
Teams still measure resilience by how many people hold a system in their heads, a measure that assumes at least one person once did, which generated code quietly stops guaranteeing.
Mechanism
Generation outruns comprehension, so a system can reach production having never been understood by anyone including its author, and the mental model that maintenance depends on was never formed rather than merely lost.
Move
Externalize the reasoning while it still exists, record decisions rather than only code, and treat any artifact nobody can explain as unfinished no matter how well it runs.
You have walked The craft end to end: the context nobody could hold, the inversion that made checking the job, the illusion that hides the error, the drift that survives review, and the absence left when none of it is done. The half of the work that got scarce was always the part that was the craft.