Prompting for Code Is Different From Prompting for Prose

A prompt that writes a poem is not the same as a prompt that ships working code. Code has hidden standards. You need to make them explicit.

30 min · Reviewed 2026

Prose Forgives. Code Does Not.

A poem about loneliness can be five lines or fifty and still work. A function that adds two numbers has to return the exact right number, in the exact right type, or the program breaks. Prompts for code have to be more precise, because code is less forgiving.

The four things every code prompt needs

Language and version (Python 3.12, TypeScript, etc.)
Inputs: types, shapes, and example values
Outputs: what to return, including edge cases
Constraints: performance, dependencies, style

A weak prompt vs. a strong prompt

Weak	Strong
Write a function to sort things	Write a TypeScript function sortByDate(items: {id: string, created: Date}[]) that returns the array sorted oldest first. Do not mutate the input.
Make me an API	Write a Node.js Express route POST /signup that takes {email, password}, hashes the password with bcrypt, and returns 201 with {id, email}.
Fix this bug	This function returns NaN when items is empty. Expected behavior is to return 0. Fix the bug and keep the signature.

The example-driven prompt template

Task: Write a function `slugify(text: string): string`. Language: TypeScript. Behavior: - Lowercase the input. - Replace spaces with hyphens. - Strip characters that are not a-z, 0-9, or hyphen. - Collapse multiple hyphens into one. Example: slugify("Hello, World!") === "hello-world" slugify(" A B ") === "a-b" Constraints: - Pure function. - No external dependencies.A prompt you could paste into any AI coding tool and get a reliable, testable function back.

What to leave out

Do not ask the AI to be creative unless you actually want variety
Do not pile on so many rules that the task becomes contradictory
Do not ask for best practices vaguely — name the practice

The best programmers write the spec before the code, whether or not an AI is helping.
— A staff engineer

The big idea: code is a contract, so a code prompt must be a contract too. Inputs, outputs, examples, constraints. Nail those, and the AI stops guessing and starts delivering.

End-of-lesson check

8 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-coding-prompting-for-code-builders

What is the main idea of "Prompting for Code Is Different From Prompting for Prose"?
1. A prompt that writes a poem is not the same as a prompt that ships working code. Code has hidden standards. You need to make them explicit.
2. Use AI as the final authority for the whole decision
3. Avoid checking the answer once it sounds polished
4. Focus only on speed instead of judgment
Which concept is most central to "Prompting for Code Is Different From Prompting for Prose"?
1. constraints
2. code prompts
3. I/O examples
4. acceptance criteria
Which use of AI fits this topic best?
1. Let the AI decide what matters without your review
2. Use the answer before checking whether it fits the situation
3. Language and version (Python 3.12, TypeScript, etc.)
4. Use the first answer without checking it
What should a careful learner remember about "Include an example"?
1. One concrete input-output pair is worth three paragraphs of description. The AI mimics examples extremely well.
2. Skip the context so the tool can guess faster
3. Treat the output as private even after sharing it online
4. Use the answer without checking the source
You want to use AI after this lesson. What is the safest next step?
1. Act immediately because the AI answer is written clearly
2. Use the AI answer as a draft, then check it against a reliable source.
3. Hide uncertainty so the final answer looks cleaner
4. Use private or sensitive details before checking permission
How should AI output about code prompts be treated?
1. As proof that no other source is needed
2. As a replacement for context, consent, or expert review
3. As a draft or helper output that still needs human judgment and verification
4. As something that becomes correct when it sounds confident
Name one way to verify an AI answer about code prompts.
Which action would help you apply "Prompting for Code Is Different From Prompting for Prose" responsibly?
1. Use the tool to avoid thinking through the tradeoff
2. Keep going even if the output conflicts with a trusted source
3. Use the first answer without checking it
4. Inputs: types, shapes, and example values

← Back to interactive lesson

Tendril · Builders · AI-Assisted Coding

Prompting for Code Is Different From Prompting for Prose

A prompt that writes a poem is not the same as a prompt that ships working code. Code has hidden standards. You need to make them explicit.

30 min · Reviewed 2026

Prose Forgives. Code Does Not.

The four things every code prompt needs

Language and version (Python 3.12, TypeScript, etc.)
Inputs: types, shapes, and example values
Outputs: what to return, including edge cases
Constraints: performance, dependencies, style

A weak prompt vs. a strong prompt

Weak	Strong
Write a function to sort things	Write a TypeScript function sortByDate(items: {id: string, created: Date}[]) that returns the array sorted oldest first. Do not mutate the input.
Make me an API	Write a Node.js Express route POST /signup that takes {email, password}, hashes the password with bcrypt, and returns 201 with {id, email}.
Fix this bug	This function returns NaN when items is empty. Expected behavior is to return 0. Fix the bug and keep the signature.

The example-driven prompt template

Task: Write a function `slugify(text: string): string`. Language: TypeScript. Behavior: - Lowercase the input. - Replace spaces with hyphens. - Strip characters that are not a-z, 0-9, or hyphen. - Collapse multiple hyphens into one. Example: slugify("Hello, World!") === "hello-world" slugify(" A B ") === "a-b" Constraints: - Pure function. - No external dependencies.A prompt you could paste into any AI coding tool and get a reliable, testable function back.

What to leave out

Do not ask the AI to be creative unless you actually want variety
Do not pile on so many rules that the task becomes contradictory
Do not ask for best practices vaguely — name the practice

The best programmers write the spec before the code, whether or not an AI is helping.
— A staff engineer

The big idea: code is a contract, so a code prompt must be a contract too. Inputs, outputs, examples, constraints. Nail those, and the AI stops guessing and starts delivering.

End-of-lesson check

8 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-coding-prompting-for-code-builders

What is the main idea of "Prompting for Code Is Different From Prompting for Prose"?
1. A prompt that writes a poem is not the same as a prompt that ships working code. Code has hidden standards. You need to make them explicit.
2. Use AI as the final authority for the whole decision
3. Avoid checking the answer once it sounds polished
4. Focus only on speed instead of judgment
Which concept is most central to "Prompting for Code Is Different From Prompting for Prose"?
1. constraints
2. code prompts
3. I/O examples
4. acceptance criteria
Which use of AI fits this topic best?
1. Let the AI decide what matters without your review
2. Use the answer before checking whether it fits the situation
3. Language and version (Python 3.12, TypeScript, etc.)
4. Use the first answer without checking it
What should a careful learner remember about "Include an example"?
1. One concrete input-output pair is worth three paragraphs of description. The AI mimics examples extremely well.
2. Skip the context so the tool can guess faster
3. Treat the output as private even after sharing it online
4. Use the answer without checking the source
You want to use AI after this lesson. What is the safest next step?
1. Act immediately because the AI answer is written clearly
2. Use the AI answer as a draft, then check it against a reliable source.
3. Hide uncertainty so the final answer looks cleaner
4. Use private or sensitive details before checking permission
How should AI output about code prompts be treated?
1. As proof that no other source is needed
2. As a replacement for context, consent, or expert review
3. As a draft or helper output that still needs human judgment and verification
4. As something that becomes correct when it sounds confident
Name one way to verify an AI answer about code prompts.
Which action would help you apply "Prompting for Code Is Different From Prompting for Prose" responsibly?
1. Use the tool to avoid thinking through the tradeoff
2. Keep going even if the output conflicts with a trusted source
3. Use the first answer without checking it
4. Inputs: types, shapes, and example values

← Back to interactive lesson