Six brief fields without which any assistant produces a template
Two people open the same assistant with the same task: the first gets a document ready for use, the second a set of general phrases. The difference is not luck but six fields one supplied and the other did not. Below are all six and which one weighs most.
Two people open the same assistant with the same task: the first gets a document ready for use, the second a set of general phrases. The difference is not luck but six fields one supplied and the other did not. Below are all six and which one weighs most.
The same tool, different results
Two people open the same ready assistant. They set the same task in substance.
The first gets a document that can go straight into work. The second gets a set of general phrases that would suit anybody.
The second concludes: "the assistant is not much good". The first does nothing, because his works.
The difference is neither the assistant nor luck. It is that the first supplied inputs and the second supplied a task.
Below are the six fields usually missing. They are universal: they work with any assistant and any task.
Why you get a template without them
The mechanics are simple and worth understanding.
A model cannot produce specifics it does not have. You asked for landing page text and did not say who for. It does not know your audience — so it writes for an average one.
The average is the template. Not because the model is lazy but because that is the only thing that can be built with no direction.
Second: a good assistant asks clarifying questions itself. But it asks them within its own frame rather than about your situation, and if you answer in monosyllables it still fills in the rest itself.
The practical conclusion: it is faster to supply the inputs at once than to accumulate them through dialogue.
The six fields
1. Who for
Not "for an audience" but specifically: who these people are, what they already know, what they have already tried, what stage they are at.
Bad: "for entrepreneurs".
Good: "owners of car repair shops in towns under half a million, no website, never done any marketing, believe customers come by themselves".
What that changes: the assistant stops explaining basics to people who know them, and vice versa.
Where to get it: from audience research, from reviews, from the questions you are asked most often.
2. What the person should do afterwards
One specific action. Not "become interested" but "leave an enquiry", "send a direct message", "book a meeting".
Without this field the material ends in nothing — and that is the most common reason a good text does not work.
3. What must not be said
The most underrated field and the strongest.
A list of prohibitions: which words not to use, which promises not to make, which subjects not to touch, which tone to avoid.
Prohibitions work harder than wishes. "Write it well" gives a mediocre result. "Do not use 'unique', 'in today's world', do not open with general reflection, do not promise sales growth" gives something entirely different.
The obligatory minimum for commercial text: do not promise income, do not promise business results, do not make medical or financial claims.
4. In what voice
Not "professionally" — that means nothing. Give an example.
The fastest way: attach two of your own texts and write "write like this". Showing a voice is easier than describing it.
What to state in words if there are no examples: sentence length, formal or informal address, whether irony is allowed, favourite turns of phrase, what never appears in your speech.
5. What already exists
Material to rely on: your data, your examples, your price list, your documents.
This is the key field for reliability. When all the material is supplied by you, there is nothing to invent — the safest mode of working there is.
The request "here is a document, analyse it against these points" is more reliable than "tell me what usually happens".
6. What should come out
Format and length. A document, a list, a table, a text of so many characters, how many sections.
Without this field you get whatever the model finds convenient, and something different every time. For processes where the result feeds onwards, that breaks the next link.
What the six brief fields look like in practice
The same task, two approaches.
The usual way: "Write a post about our services."
What comes out: smooth text for nobody, with a list of advantages and a call to get in touch.
With the six fields:
Who for: owners of car repair shops, no website, believe they do not need one. The real pain is customers ringing with the same question about prices.
The action: send a direct message with the word "prices".
Not allowed: the words "unique", "innovative", "in today's world". Do not promise sales growth. Do not open with general reflection.
Voice: short sentences, formal address, no irony. An example of my own text is attached.
Material: here are ten real customer questions.
Format: up to 900 characters, short paragraphs, no list of advantages.
The second version is several times longer. But it is written once and reused thereafter: only the material changes.
A template instead of doing it every time
Hence a practical technique that saves the most.
Assemble the brief once and keep it. The first four fields — who for, the action, the prohibitions, the voice — change rarely for you. Material and format are substituted per task.
Every new request then starts by pasting that template in.
If you work with clients, a brief for each. One document per client: his audience, his voice, his prohibitions. A second job for the same client costs you several times less than the first.
And that is also what becomes a knowledge base if you are assembling an assistant for a company.
How to check the inputs are enough
A simple test before sending.
Read your request and ask: could a competitor put his own name in here and have it all remain true?
If yes, the inputs are insufficient and the result will be a template. Not because the request is bad but because there is nothing of yours in it.
A second test: is there at least one of the six fields in the request that others definitely do not have? Usually it is the material — your data, your examples, your clients' questions.
What the platform offers for a brief
Over 200 ready assistants across 14 categories. Each has its own frame and its own clarifying questions — which is exactly what removes part of the work on inputs. But the material and the prohibitions are still yours.
Refinement modes: assessing the work, harsh assessment, focus group, criteria-based refinement. A useful technique is to give the first result to the harsh assessment and see what the model itself thinks was missing. That is usually where the missing field turns up.
Prompt Master in Gen AI — the "Improve prompt" button rewrites the phrasing to a method. It does not replace inputs but helps with form.
“Freelancer positioning” — if the first field, "who for", is not filled in at all: a chain of seven assistants gives a niche and a client avatar. $9, included in Basic.
Your own assistant (Creator plan) — if the request recurs, the inputs are fixed once in its instructions and only the material changes.
Registration is free and opens three days of full Basic access.
Where to start today
One action. Take the last unsuccessful result — the one that came out as a template — and check your request against the six fields.
Usually two are filled in: what needs doing and in what format. The other four the model filled in itself.
Add the third field — the list of prohibitions. That one field usually makes the most noticeable difference, and filling it takes a minute.
Where the brief is already built into the tool
Assistants do not require a brief from scratch — they assemble it themselves. In the catalogue, Basic plan, each is set up to ask the right questions: the Personality unpacker leads through a dialogue, the Landing page architect first analyses a reference and clarifies the inputs, the Creative social packager begins directly by gathering data about you.
That is an answer to the article's problem from the other side: instead of learning to write a brief, you can take a role that extracts one.
If the brief has to be your own. The “Personal AI bureau” express course ($9, Creator plan and above) — how to design an assistant for a task: a library of 100+ industrial instructions across six directions, each built to a single architecture, protected against copying and ready to sell as a product.
A mega prompt as a ready frame. The express courses on Claude and on Gemini ($9 each, Creator plan and above) give a universal template: role, context, task, response structure, constraints, success criteria. Fill the blocks in for your task and you get a consultant-level prompt. Understand the structure once and you stop hunting for good prompts in chats.
Where the phrasings themselves come from
Six fields are what to supply. A separate question is how to phrase it so the model understands.
The “Basic prompting” course devotes two lessons to this.
Tone of voice — how to set the voice so the result sounds like you rather than like averaged text.
Role prompts — how to assign a model a role so it holds it to the end rather than sliding into a general tone after three paragraphs.
And one more lesson of the course, “Cascading prompts”. It explains what to do when the six fields are filled and the task is still too large: break it into a chain where each step solves its own sub-question and passes the result on.
The phrasing from the lesson: you lead the model along a clear plan, like a scriptwriter.
The “Basic prompting” course — Basic plan.
What to take on the platform for this task
- Basic prompting — Tone of Voice, role prompts and cascading prompts; a course, Basic plan
- Prompt engineering: a foundational course for working with all AI models — $9, Basic plan and above
- Personal AI bureau — $9, Creator plan and above
- Landing page architect — an assistant that takes the brief inputs (from Basic)
What to read next
One assistant against three — what to do when one is not enough.
What prompt engineering is — the fundamentals of setting a task.
How Claude differs from other models — why structure works.
How to build your own AI assistant — how to fix the fields once.
AI agents: what they are and what they pay — the full guide to agents, with market rates.