What AI models work with these prompts?

The prompts are designed to work with all major language models including ChatGPT (GPT-4), Claude, Gemini, and Llama. Most prompts use universal techniques that transfer across models without modification.

Can I customize the prompts for my use case?

Yes. Every prompt includes editable placeholders in brackets so you can tailor it to your industry, tone, or specific task. The prompts are starting points, not rigid scripts.

Are these prompts free to use?

All prompts on this page are completely free. Copy and use them however you like — in your personal projects, at work, or to build products. No attribution required.

How are the prompts organized?

Prompts are organized by use case: writing, coding, research, analysis, brainstorming, and more. You can filter by category or search by keyword to find the right prompt quickly.

What makes a good AI prompt?

The best prompts specify a role, provide context, define the output format, and include constraints. Our prompts follow these principles and include reasoning chains that help models produce more accurate, structured results.

AI & Productivity

50 ChatGPT Prompts That Actually Work

Most ChatGPT prompts produce generic slop because they are too vague. The difference between a mediocre prompt and a great one is structure. Here are 50 prompts that consistently deliver useful output, organized by category, with the formula that makes each one work.

Why Most Prompts Fail

Typing "write me an email" into ChatGPT is like telling a chef "make me food." You will get something, but probably not what you wanted. The three most common prompt failures:

Too vague: No context about audience, tone, or purpose. The model guesses and usually guesses wrong.
No format instruction: You wanted bullet points but got five paragraphs. You wanted a table but got prose. Always specify the output format.
No constraints: Without word limits, length guidelines, or scope boundaries, the model defaults to exhaustive (and exhausting) responses.

The Prompt Formula

Every effective prompt contains up to five elements. You do not need all five every time, but the more you include, the better the output:

Role: Who should the AI act as? ("You are a senior copywriter specializing in B2B SaaS.")
Context: What is the situation? ("I am launching a new feature for project managers.")
Task: What exactly should it produce? ("Write 5 email subject lines.")
Format: How should the output look? ("Return as a numbered list with character count for each.")
Constraints: What are the boundaries? ("Under 60 characters each, no exclamation marks, professional tone.")

Writing Prompts

1. Rewrite for Clarity

Rewrite the following paragraph for a general audience at an 8th-grade reading level. Keep the same meaning but eliminate jargon and shorten sentences to 20 words max. [paste text]

Why it works: Specifies audience, reading level, and sentence-length constraint.

2. Blog Post Outline

You are a content strategist. Create a detailed outline for a 1,500-word blog post titled "[TITLE]". Include H2 and H3 headings, estimated word count per section, and a one-sentence summary for each section. Target audience: [AUDIENCE].

Why it works: Role + task + format + constraints (word count, heading levels).

3. Email Reply Draft

Draft a professional email reply to the message below. Tone: friendly but firm. Goal: decline the request without damaging the relationship. Keep it under 100 words. [paste email]

4. Product Description

Write a product description for [PRODUCT] in exactly 3 sentences. Sentence 1: what it is. Sentence 2: the main benefit. Sentence 3: who it is for. Tone: conversational, not salesy.

5. Social Media Caption

Write 5 Instagram captions for a post about [TOPIC]. Each caption: max 150 characters, include one emoji, end with a question to boost engagement. Tone: authentic, not corporate.

Coding Prompts

6. Debug This Code

I have a bug in the following [LANGUAGE] code. The expected behavior is [X] but the actual behavior is [Y]. Identify the bug, explain why it happens, and provide the corrected code. [paste code]

7. Explain Code to a Junior

Explain the following code to a junior developer who has 6 months of experience. Use simple language. After the explanation, suggest one improvement. [paste code]

8. Write Unit Tests

Write unit tests for the following function using [TEST FRAMEWORK]. Cover: happy path, edge cases (empty input, null, boundary values), and one error case. Include descriptive test names. [paste function]

9. SQL Query Builder

I have tables: [describe schema]. Write a SQL query that [describes goal]. Use CTEs for readability. Add comments explaining each CTE. Optimize for PostgreSQL.

10. Code Review

You are a senior [LANGUAGE] developer conducting a code review. Review the following pull request for: bugs, performance issues, readability, and security concerns. Format as a numbered list with severity (critical/warning/suggestion). [paste code]

Business Prompts

11. SWOT Analysis

Create a SWOT analysis for [BUSINESS/PRODUCT]. Format as a 2x2 table. Include 4-5 bullet points per quadrant. Base the analysis on [INDUSTRY] market conditions in 2026.

12. Meeting Agenda

Create a 30-minute meeting agenda for [TOPIC]. Include: objective, attendees needed, 3-4 agenda items with time allocations, pre-read materials needed, and expected decisions/outcomes.

13. Competitor Analysis

Compare [PRODUCT A] vs [PRODUCT B] vs [PRODUCT C] across these dimensions: pricing, key features, target audience, and main weakness. Return as a markdown table. Add a "bottom line" recommendation at the end.

14. Elevator Pitch

Write a 30-second elevator pitch for [COMPANY/PRODUCT]. Structure: hook (problem), solution, traction/proof, ask. Max 75 words. Conversational tone, no buzzwords.

15. OKR Generator

Generate 3 quarterly OKRs for a [ROLE] at a [COMPANY TYPE]. Each OKR: 1 objective (qualitative, inspiring) with 3 key results (quantitative, measurable). Ensure key results are ambitious but achievable.

Creative Prompts

16. Story Starter

Write the opening 200 words of a short story in the genre of [GENRE]. Setting: [SETTING]. The opening must introduce a mystery or tension in the first two sentences. Write in present tense, first person.

17. Brainstorm Ideas

Generate 20 creative ideas for [GOAL]. Use the SCAMPER method (Substitute, Combine, Adapt, Modify, Put to other uses, Eliminate, Reverse). Label which SCAMPER technique each idea uses. Include at least 3 unconventional ideas.

18. Name Generator

Generate 15 name ideas for a [TYPE OF BUSINESS]. Requirements: easy to spell, available as a .com domain (check plausibility), memorable, and conveys [FEELING/VALUE]. Group into: literal names, abstract names, and compound words.

19. Dialogue Writer

Write a dialogue between [CHARACTER A: brief description] and [CHARACTER B: brief description] about [TOPIC]. The conversation should reveal a hidden conflict. 300-400 words. Include stage directions in brackets.

20. Metaphor Machine

Explain [COMPLEX CONCEPT] using 5 different metaphors. Each metaphor should be drawn from a different domain: sports, cooking, nature, music, and construction. Include a 1-sentence explanation of why each metaphor works.

Research Prompts

21. Literature Review

Summarize the current state of research on [TOPIC]. Cover: key findings from the last 5 years, major debates, methodological trends, and gaps in the literature. Format as 5 paragraphs with citations to specific studies where possible.

22. Devil's Advocate

I believe [POSITION]. Act as a skilled debater arguing the opposite. Present the 5 strongest counterarguments, each with supporting evidence. Then rate each counterargument's strength from 1-10 and explain your rating.

23. Explain Like I Am 5

Explain [COMPLEX TOPIC] as if I am 5 years old. Use only simple words (under 3 syllables). Include one analogy using something a child would understand. Max 100 words.

24. Data Interpreter

Analyze the following data and provide: 3 key insights, 2 potential concerns, and 1 recommended action. Present insights as bullet points with confidence level (high/medium/low). [paste data]

25. Source Evaluator

Evaluate the credibility of the following claim: "[CLAIM]". Assess: source reliability, potential biases, sample size (if applicable), peer review status, and replication. Rate overall credibility 1-10 with justification.

Advanced Techniques

Once you are comfortable with the formula, level up with these techniques:

Chain prompting: Break complex tasks into steps. First prompt generates an outline, second prompt expands each section, third prompt edits for tone. Each step builds on the previous output.
Few-shot examples: Show the model 2-3 examples of the output you want before asking it to generate new ones. "Here are two product descriptions I like: [examples]. Now write one for [new product] in the same style."
System instructions (custom GPTs): Set persistent context that applies to every message. Useful for establishing tone, domain expertise, and output format once rather than repeating it.
Self-critique: After generating output, add: "Now critique this output. List 3 weaknesses and rewrite to address them." The model's second pass is almost always stronger.
Specificity stacking: If the first output is too generic, do not start over. Add constraints: "Make it more specific to [industry]. Replace generic examples with ones from [domain]. Cut length by 30%."

Want more prompts?

Browse our full prompt library with categories, filters, and one-click copy.

Browse our full prompt library →