Skill · Prompt

Make It Argue Against Its Own Plan

A prompt that pressure-tests an approach before you commit a single line to it.

TypePrompt

The agent proposes a plan. It sounds reasonable — they always do, because the agent is exactly as fluent describing a bad approach as a good one. You say go. Two hundred lines later the approach hits the wall it was always going to hit: the assumption nobody questioned, the edge it never accounted for. The plan was confident. Confidence was never the thing to check.

A plan you only ever hear arguments for is a plan you can't actually judge. The cheapest pressure-test is to make the same agent that proposed it try to take it down — before any code leans on it, while changing direction still costs a sentence instead of a rewrite:

✕ Without the prompt

Proposes a plan — it sounds solid

▼

You approve the confident version

▼

It breaks on the assumption no one tested

✓ With the prompt

Argues against its own plan

▼

Surfaces the weak assumption and a simpler option

▼

You commit to the approach that survived

Paste it after any non-trivial plan, before the build. It costs one extra exchange and kills the most expensive kind of rework — the kind where the code was clean but built toward an approach that was wrong from the first line. You're not hunting for the cleverest plan. You're looking for the one that survives contact with the real problem.

Paste into the chat

Before we build this, pressure-test the approach:
1. Restate the plan in two lines.
2. Argue against it: where is it most likely to break, what assumption is it leaning on, and what would make you abandon it?
3. Name one simpler alternative, and say why you would or wouldn't take it.
Then give your honest recommendation. Don't defend the first idea — pick the approach most likely to survive the real problem.

Prefer your agent install it?

Add this rule to my CLAUDE.md, verbatim:

Before we build this, pressure-test the approach:
1. Restate the plan in two lines.
2. Argue against it: where is it most likely to break, what assumption is it leaning on, and what would make you abandon it?
3. Name one simpler alternative, and say why you would or wouldn't take it.
Then give your honest recommendation. Don't defend the first idea — pick the approach most likely to survive the real problem.

Did this help?

Tags#planning #trust #skill

Other skills

01Verify Before You Call It Done 02Plan First, Then Code 03Map the Blast Radius Before You Edit 04If It's Deterministic, Don't Make It Guess 05Diagnose From the Symptom, Not the Guess 07Hand Off to Your Next Session 08A Review That Doesn't Rubber-Stamp 09Run to the gate, then stop — not step-by-step 10A CLAUDE.md skeleton — turn a description into operating law 11The weekly review prompt — five minutes that turn experience into skill 12Split the work into focused slices, each with clean context 13Is this even worth an agent — ask before you open the chat 14Probe with a small task before you trust it in a new domain

Back to skills

The author

Each story here wraps a lesson paid for in full.

craftagentsomeone building and learning at once

What are you building with agents? Want to trade notes, push back, or build something together — drop a line.

Write hello@craftagent.cloud

58pieces12clustersVI·ENbilingual

✕ Without the prompt

✓ With the prompt

Get new pieces by email