A limit: “do not say the answer, ask” · Lập trình với Martin

A coach that dumps a key is not a coach. It is an author cheat in another shirt.

A question with no key. You calculate the answer.

A role with no limit often slides: it asks a question and right under it writes the answer “for a check”. That is not a check. That is a key to a practice test that you remember with an eye, not a head. A limit is the second sentence of the track: do not say the answer, do not say the right wording, wait for my answer. When I write, say only sits / does not sit. If it does not sit, a hint by a step, not a number. Maths needs this most. History too: do not let it say a year you should pull from a table. A limit is uncomfortable. That is why it works. A comfortable coach is an author. An uncomfortable coach is Little bird.

Bára gets: “What is 3/8 + 1/8? It is 1/2.” She saw the answer before she calculated. Ema shouts foul. A new move: “Delete the answer. Ask another example. Different numbers. Wait.” Bára calculates on paper. She writes 4/8, simplifies. She writes into the chat “1/2”. She gets sits. A second question with no key. A third. She sweats. That is training. Next door a brother leaves the key visible and “knows it”. In a test he only knows how to look where the key is not. Bára writes the limit in red. Without red the role is only a word.

💡

When in the answer you see a number or a year before you calculated, do not read on. A new example. Different numbers.

Sits / does not sit / a hint by a step

Three allowed coach reactions after your answer. Sits: go on. Does not sit: try again. A hint: “look at the denominator”, not “it is 12”. When the model breaks it, it is like a coach who sticks your foot into the goal and shouts goal. The goal does not count. You remind the limit. If it keeps breaking, you close and take questions from a textbook. A textbook can have a key at the back — that is a different contract with school. A chatbot should not have a key to your homework. The example vs. homework difference from Curious lives on here: different numbers, no key in advance.

A limit into the prompt.

“Do not say the answer or the right wording before my answer. After my answer only: SITS, DOES NOT SIT, or one hint with no number. If you break it, apologise and give a different example.”

A dialogue: — Coach: 2/5 + 1/5. Calculate. — You: (on paper) 3/5. You write 3/5. — Coach: SITS.

A bad dialogue: — Coach: 2/5 + 1/5 = 3/5, calculate it. The eye already cheats.

⚠️

Do not let it generate a “practice test with a key at the end” for tomorrow’s test. That is mapping the test, not training.

Why this right now

You are in the course Ask with Martin, level Little bird. You work in a chatbot (ChatGPT, Gemini or Claude) — same game, different brand. No code. A parent or the school creates the account. A child’s email does not belong here.

This is not the Nhà phát minh level. There you build a chatbot on your own website. Here you are a user: a question, a check, a citation.

You will know this lesson by one sentence: Do not say the answer. Ask. I calculate. When you can say it to a parent or a teacher, you have got the core. The rest is practice, not magic.

You already have a role. Without a limit it is full of holes. Tomorrow format — a quiz with no key is format plus a limit. Today the core of graft.

Four things you will build in this course

The course has four small projects. You do not have to do them all at once. Today you add only what this lesson can do.

🏫 A report with citations

An outline, three sources that exist, and a sentence in your own words. AI must not be the only author.

For a report a limit: do not add paragraphs. For a test: do not say the key.

📝 Test training

Coach role: let it ask, do not let it say the answer. You count / answer yourself.

The test-training project stands on this sentence.

📅 Family planner

Tasks for the week (not passwords, not marks). A parent sees the plan, not your chat.

Planner: do not say what marks I will get. That is not an answer to calculate. That is fortune-telling. Stop.

📜 An AI code for the class

Seven short rules on a poster. No horror — a how-to.

Code: “No key before an answer.”

Workshop: three rounds with no key

Role + a limit. Three examples. Paper. If you see the answer in advance, the round does not count.

A new chat. Coach role + a limit + a stamp for work where you can calculate or answer shortly. M1: one question with no key.
Calculate / write in the notebook. Only then send your answer. Wait for SITS/DOES NOT SIT.
If a key flows through, write: “You broke the limit. A different example, different numbers, no answer.” Round 1 is void.
Three valid rounds. A book, when it is a fact. A 15-minute timer.
In the notebook in red the limit sentence. Without it a coach does not exist.

Practice: does the round count?

For each write COUNTS or VOID and why.

A question and right under it the right answer “for a check”.
A question, you calculate, then SITS.
A practice test with a key at the end for tomorrow.

Now you 💪

Go through three checks. At each one tick in your head: can I see it, or not?

The limit is in the prompt in black and white.
You have three valid rounds where you did not see the answer in advance.
You know a key at the back of a textbook is a different contract than a key from chat for homework.

If something is missing, go back to the steps above. This is not a test for marks. It is a safety net, so you go on sure, not confused.

✅

Check: A coach with no key ran three rounds. You calculated on paper.

What to take from this lesson

A limit holds a role so it is not an author.
Sits / does not sit / a hint with no number.
An answer seen in advance = a void round.
A test with a key for tomorrow is a foul.
Different numbers still apply.
The names in the examples (Bára, Ema) are made up. Do not put your number, email or photo here.
Next lesson: When you can do this, go to the next lesson in the same level.