Skip to lesson
Exit
AI-assisted review, testing & anti-patterns1 / 3

1 min lesson

How they cluster

In "How they cluster", respond to "What single sentence captures why every anti-pattern in the taxonomy is recoverable?" Give the answer first and the evidence second.

Step 1 of 3

How they clusterthree root causes

Process collapse: vibe-merges, mega-diffs, prompt-and-prayEpistemics: fabricated confidence, hidden generated code, context rotThe quality drop that happens when one chat accumulates unrelated tasks or tangents, burying the signal the model needs in a haystack of stale context. Press Enter for the full definition.Security & incentives: secrets/injection, excessive permissions, volume-as-success

Notice the through-line. Nearly every guardrail in the right column is something we already covered: the well-formed PR, the layered review with a non-collapsible human gate, intent-asserting tests, least privilege, outcome metrics over volume. The anti-patterns aren't exotic. They're what you get when you skip the disciplines. Your job in the field is to spot which discipline a team dropped, then restore it.

Learn more

Full explanation

The canonical prompt anti-pattern

The canonical prompt anti-patternwhere prompt-and-pray actually starts

Most prompt-and-pray messes begin with the same move: a vague prompt on the most expensive model. "Make the app better" on a frontier model, expecting horsepower to substitute for direction. With no concrete target the agent makes blind, scattershot edits, which burns tokens, then burns more tokens reading and fixing the bad code it just wrote: a compounding blind edits → read/fix → more tokens loop that drives cost up and quality down at the same time. Power on a fuzzy task is the worst of both worlds.

Vague (drives the loop)
"Make the app better."
Scoped (names surface, change, constraints)
"Under the shop page, add an in-stock filter next to the existing tag and sort controls. Preserve URL query params and update tests if needed."
Vague (drives the loop)
Any model, no success check
Scoped (names surface, change, constraints)
Match model power to task; in plan mode for anything non-trivial

Specificity, not model price, is what stops the blind-edit loop. The fix is a concrete success check, plan mode first and a fresh agent when context bloats.

Learn more

Optional practice

Practice: How they cluster

QWhy is 'make the app better' on the most expensive model the canonical anti-pattern, and what's the fix?