Prompting for general tasks: common mistakes and how to fix the answer

Hello again. Let’s talk about prompting for everyday tasks: what to put in, what people leave out, and how to fix a bad answer without starting over.

By the end of this article, you will know the four parts of a prompt that works. You will be able to spot the seven mistakes that cause most bad answers. And you will be able to read a wrong answer, find the missing piece in your prompt, and fix that one piece instead of starting over.

You paste a report into the chat box and type “summarise this.” Back come five bland bullet points. They could describe almost any report. There is nothing you can use in the meeting you are prepping for. Or you type “write me an email to my landlord” and get a stiff, generic note. It does not mention the broken heater, the date, or the fact that you have already emailed twice. The answer disappoints. The easy conclusion is that the model is not very good. It usually is. The prompt just did not say enough for any answer to be the right one.

“Summarise this” never told the model who the summary is for. It never said how long it should be, or what you care about. “Write me an email” never said what the email is about, in what tone, or with what outcome. A person handed the same instruction would ask follow-up questions. The model does not always ask. It guesses, and an average guess is what you got.

This is the Description skill from the AI fluency framework: saying clearly what you want so the model can act on it (more on the wider framework in the 4D framework for working with AI). The other half is Discernment: judging whether the answer is right before you use it. Most everyday frustration with AI lives in the gap between those two. This is a field guide to closing that gap. It covers what a workable prompt is made of, the handful of mistakes that produce most bad answers, and how to repair a response without starting from scratch.

Anatomy

What a prompt that works is made of

A prompt the model can act on has four parts. Not every prompt needs all four. But when an answer disappoints, the missing piece is almost always one of these.

The four parts of a workable prompt. Role and context tell the model where it stands; the task says what to do; constraints and format set the shape; an example shows the target. Most weak prompts are missing two of the four.

Take the weak prompt and rebuild it. “Summarise this” becomes: “You’re helping me prep for a budget meeting (role and context). Summarise the attached report (the task) as five bullets, each under fifteen words, focused only on cost changes and risks (constraints and format). Here’s the style I want: ‘Cloud spend up 18%, driven by the new logging pipeline’ (example).” The model now has somewhere to stand. The summary comes back about the thing you actually care about. It comes in a shape you can paste straight into your notes.

The same rebuild works on the email. Give it the context (broken heater, two prior emails, you want a repair date). Give it the task (a firm but polite follow-up). Give it the format (short, three sentences), and a line showing the tone you want. Each part removes one thing the model would otherwise have to guess, and that is the only reason to make the prompt longer.

The usual mistakes

Seven ways prompts go wrong

Across Anthropic’s prompt engineering overview and its interactive tutorial, the same techniques keep showing up as the ones with the most impact: be clear, give context, show examples, let the model reason, name the format, give a role. The common mistakes are just those techniques missing. Here are the seven that produce most bad answers, each with a one-line fix.

Vague instruction. “Make it better,” but better how? Fix: say exactly what you want, the way you would brief a new colleague with no background.
Missing context. The model cannot know your audience, your limits, or why you are asking. Fix: state the why, the who, and the limits up front.
No example. You describe the output instead of showing it. Fix: paste one or two samples of the shape you want.
No format named. You leave the shape to chance and get a wall of prose. Fix: name it, whether a table, three bullets, or a 200-word summary.
Too many things at once. One prompt asks for a summary, a translation, and a plan. Each comes out thin. Fix: do one job per prompt, or number the asks.
A reasoning question with no room to reason. You ask for a judgement but want it instantly. Fix: ask the model to think through the steps before giving the answer. This is called chain-of-thought, and Anthropic recommends it for anything analytical.
Assuming it remembers. You refer to “the document from earlier” that it never actually saw, or that scrolled out of context. Fix: re-paste the thing you are referring to instead of pointing at it.

The seventh mistake is not in the prompt at all: trusting the first answer without checking it. That is the Discernment side of the AI fluency framework. An answer can read fluent and confident and still be wrong.

Fixing the answer, not starting over

When the response is wrong, read the symptom

When an answer comes back wrong, the instinct is to rewrite the whole prompt and try again. Usually you do not need to. The shape of the failure tells you which part of your prompt was thin. Read the symptom, trace it to the cause, change one thing.

Debugging an answer by its symptom. The way an answer is wrong points back to what your prompt left out. Fix the one missing part rather than rewriting everything.

Symptom in the answer	Likely cause in your prompt	The fix
Too generic, could be about anything	No context or example given	Add context and one example of the output you want
Made something up (a fact, a quote, a citation)	Asked beyond what the model reliably knows	Give it the source to work from; tell it to say “I don’t know” rather than guess
Wrong format: prose when you wanted a table	No format named	Specify the shape: a table with these columns, three bullets, JSON
Far too long or too short	No length set	Set a concrete limit: “in 200 words”, “exactly five bullets”
Ignored part of what you asked	Too many asks crammed into one sentence	Number the asks, or split them into separate prompts
Right facts, wrong tone	No role or voice set	Give it a role: “you’re a careful copy editor”, “write it warmly”
Refused or hedged when it shouldn’t have	A constraint read as off-limits, or the ask was ambiguous	Rephrase the constraint and say why you’re asking and what it’s for

The made-something-up row costs people the most. A model fills gaps with plausible text. So a question it cannot answer from what it knows produces a confident invention rather than a blank. Give it the source to quote from, and clearly allow “I don’t know.” That turns a guess back into an honest gap.

Don’t restart, iterate

Steering in follow-ups

You almost never need a fresh prompt. The chat already has the context, the draft, and your earlier instructions. Correcting in place is faster and keeps all of that. When an answer is 80% there, say what to change: “Good, but tighten it to three bullets and keep the second point exactly as is.” Or: “Same email, but drop the apology in the first line and make the deadline firmer.”

This is the Description and Discernment loop running live. You judge the answer (Discernment). You describe the adjustment (Description). The answer improves, and you judge again. A few short turns usually beat one long prompt. You are reacting to a real output instead of trying to predict every requirement in advance.

A short checklist

Before you hit enter

A quick pass before you send, and a habit for after. Most of the seven mistakes get caught here.

Is the ask clear? Would a new colleague know exactly what to do from this sentence alone?
Did I give context? Who it is for, why, and any limits the model cannot know.
Is there an example? For anything where format or style matters, show one.
Did I name the format and length? The shape (table, bullets, prose) and the size.
One job or many? If it is many, number them or split them.
Will I actually check the answer? Decide before you read it that you will verify the facts, not just the fluency.

That last item is the Discernment habit. It matters most when you will act on the output. The case for checking what you ship, rather than trusting a confident-looking result, is the whole argument of vibe coding versus engineering: the output looking right is not the same as it being right. If you want to get better at this steadily rather than picking it up by chance, becoming AI-fluent: a learner’s playbook is the longer path. Anthropic’s own foundations course frames the same two skills as the base everything else builds on.

Recap

The goal of this article was to help you write prompts that work, and fix the ones that do not, without starting over. The main points:

A workable prompt has four parts: role and context, the task, constraints and format, and an example. A missing example is the most common gap.
Most bad answers come from seven mistakes: vague ask, missing context, no example, no format, too many asks, no room to reason, and assuming the model remembers.
When an answer is wrong, read the symptom. It points to the one part of your prompt that was thin. Fix that part only.
Correct in follow-ups instead of rewriting. Judge the answer, describe the change, judge again.
Always check the answer before you use it. A confident tone is not proof.

Good prompting is the plain habit of saying what you want clearly, showing an example when shape matters, reading the answer back carefully, and adjusting in place. Do that, and most of the answers that used to disappoint stop disappointing.

I hope this writing helps you. Thanks for reading, and feel free to share it with someone who is still typing “summarise this.”