How is this different from a "top 100 ChatGPT prompts" blog post?

Those lists are mostly scraped or LLM-generated filler. Every template here is something we actually used with ChatGPT, Claude, or Gemini ourselves, and kept because it gave better output than the obvious one-line version of the same ask. The library is structured: each prompt has explicit {placeholders}, a category, a one-line "what it does" description, and the model we found handles it most reliably (or "any" if all three work). No infinite scroll, no "best prompts of 2025" recycling.

What do the {placeholders} do?

Many prompts have variables like {topic}, {audience}, {tone}, {code}. The tool parses every placeholder out of the prompt body and renders an input box for each. As you type, the preview pane assembles the final prompt live. When you copy, you get the filled version — not the template — ready to paste into ChatGPT, Claude, or Gemini. If a prompt has no placeholders the tool tells you to copy as-is.

Why label templates as "best for ChatGPT" vs Claude vs Gemini?

From our own A/B testing: ChatGPT handles structured outputs and numbered-list instructions most reliably; Claude is more careful on long-context analysis and code review; Gemini wins on multilingual and search-style factual lookups. Templates where all three work are labeled "any" — we do not try to push you to one model out of brand loyalty. Most templates work on any modern model; the label is a hint, not a hard requirement.

Do you call any AI API? Is my input sent anywhere?

No. The entire library runs in your browser. The 200+ templates ship as a JavaScript array in the page. Search, category filter, placeholder extraction, and prompt assembly all run client-side. We do not call OpenAI, Anthropic, or Google. We do not log what you typed into the placeholder inputs. The copy-to-clipboard uses the native browser API.

Can I contribute or suggest a template?

Yes — the library is part of an open project. The data lives as a plain TypeScript array in the component source. If you have a template that earns its keep (i.e. you have used it 5+ times), open an issue or PR describing the template, the placeholders, the category, and one real example of input + output. We reject generic "act as an expert in X" templates because they do not add value over the obvious phrasing.

Why 12 categories specifically?

They map to the 12 use cases we see prompts repeatedly clustered around in real usage: writing, coding, learning, marketing, job search, SEO, creative, translation, docs, data analysis, role play, productivity. We deliberately did not split into 30 fine-grained categories because the search bar already does the long-tail filtering — categories are for browsing intent, not exhaustive taxonomy.

Is "prompt engineering" still a thing in 2026?

Less than it was in 2023 — modern models handle vague asks much better than the GPT-3 era. But specific structural patterns still matter: numbered constraints, role + audience + format framing, explicit "what to do when uncertain", and few-shot examples in the right places. Most templates here encode one of those patterns. So the answer is: less ceremony, more structural discipline. The templates are templates because the structural part is what is worth saving.

Stop rewriting the same "summarize this meeting" prompt every Monday

You paste a raw transcript into Claude 3 times a week and retype the same framing each time. Open the docs category, grab the meeting-notes template, fill {transcript} and {audience}, copy the assembled prompt. It already encodes the structure that worked for you: decisions first, owners and dates next, open questions last. Ten seconds instead of two minutes of retyping, and the output shape stays consistent every week.

Brief a junior writer who keeps getting the tone wrong

Your blog drafts come back too formal. Pick the writing template with {topic}, {audience}, {tone}, fill tone as "plain, second person, short sentences", and hand the assembled prompt to the writer to paste into ChatGPT. Because the role and format framing is baked in, three people on the team produce drafts in the same voice instead of three different interpretations of "make it casual".

Translate product copy into 5 languages without losing the CTA

You have 20 button labels and microcopy strings to localize. Use the translation template marked best-for-Gemini, fill {source_text} and {target_language}, and note "keep it under 4 words, preserve the call to action". The label tells you Gemini handles the multilingual lookup most reliably, so you skip the round of testing which model to use and go straight to filling 5 language slots.

Turn a messy bug report into a structured repro for the dev team

A user pastes a 3-paragraph rant about something broken. Drop it into the coding template, fill {raw_report}, and the assembled prompt asks ChatGPT to extract steps to reproduce, expected vs actual, and the environment. You get a clean ticket body in one copy-paste instead of interrogating the prompt structure from scratch each time a vague report lands in the inbox.

200+ Prompt Templates for ChatGPT, Claude & Gemini

200+ prompt templates for ChatGPT, Claude, Gemini — copy-paste, browse by use case.

Runs locally
Category Text
Best for Removing repetitive cleanup work from everyday writing and operations.

341 templates

Blog post outline

Drops a clean H2/H3 outline you can hand to a writer, with hook + CTA already structured.

Best for: ChatGPT / Claude / Gemini

Fill placeholders

Preview

You are an experienced editor. Write a detailed outline for a blog post titled "{title}" aimed at {audience}. The post should be ~{word_count} words. Include: a hook intro, 5-7 H2 sections with 2-3 bullet sub-points each, and a closing CTA. Tone: {tone}.

What this tool does

A curated library of 200+ AI prompt templates we actually use ourselves with ChatGPT, Claude, and Gemini — not a scraped "top 100 prompts" list. Browse by 12 use cases: writing, coding, learning, marketing, job search, SEO, creative, translation, docs, data analysis, role play, productivity. Each template has a clear title, the full prompt body with {placeholders} you can fill in, a one-line description, and the model that handles it best in our own testing. The placeholder form parses every {variable} out of the template, gives you an input box for each, and assembles the final prompt live — no AI API call, no tracking, all client-side. Search is ranked: 3 points for a title hit, 2 for a description hit, 1 for the prompt body. One click copies the assembled prompt straight to your clipboard. The library leans into Claude for long-context analysis, ChatGPT for structured lists, Gemini for multilingual lookups — and a lot of templates are explicitly marked "works on all 3". No fluff categories, no padded "best prompts" filler. Just 200+ templates that pull their weight.

Tool details

Input: Files + Text; The page exposes text boxes, numeric controls, file pickers, or structured inputs depending on the tool.
Output: Live result + Copy + Preview; The result area focuses on usable output, with copy, download, or preview actions when supported.
Privacy: Browser-side processing; The main tool logic does not call an external API, so inputs normally stay in the current tab.
Save / share: No account required; Open the page and use it; whether results survive refresh depends on the tool.
Performance budget: Initial JS <= 70 KB; No WASM budget is declared, keeping the tool quick to open on mobile.
Best fit: Text · Content Creator; Category and role tags drive related tools, internal links, and quick fit checks.

How to use

1. Input

Paste or drop your content into the tool panel.
2. Process

Click the button. All processing is local in your browser.
3. Copy / Download

Copy the result or download to disk in one click.

How Prompt Template Library fits into your work

Use it to clean, compare, reshape, or extract plain text before it goes into a document, CMS, spreadsheet, or prompt.

Text jobs

Removing repetitive cleanup work from everyday writing and operations.
Making text easier to compare, paste, publish, or feed into another tool.
Working with content locally when the text is private or unfinished.

Text checks

Scan for unintended whitespace, duplicate lines, and lost punctuation.
For long text, test the first few lines before applying the whole change.
Copy the final output only after checking the preview.

Good next steps

These links move the current task into a more complete workflow.

Real-world use cases

Stop rewriting the same "summarize this meeting" prompt every Monday
You paste a raw transcript into Claude 3 times a week and retype the same framing each time. Open the docs category, grab the meeting-notes template, fill {transcript} and {audience}, copy the assembled prompt. It already encodes the structure that worked for you: decisions first, owners and dates next, open questions last. Ten seconds instead of two minutes of retyping, and the output shape stays consistent every week.
Brief a junior writer who keeps getting the tone wrong
Your blog drafts come back too formal. Pick the writing template with {topic}, {audience}, {tone}, fill tone as "plain, second person, short sentences", and hand the assembled prompt to the writer to paste into ChatGPT. Because the role and format framing is baked in, three people on the team produce drafts in the same voice instead of three different interpretations of "make it casual".
Translate product copy into 5 languages without losing the CTA
You have 20 button labels and microcopy strings to localize. Use the translation template marked best-for-Gemini, fill {source_text} and {target_language}, and note "keep it under 4 words, preserve the call to action". The label tells you Gemini handles the multilingual lookup most reliably, so you skip the round of testing which model to use and go straight to filling 5 language slots.
Turn a messy bug report into a structured repro for the dev team
A user pastes a 3-paragraph rant about something broken. Drop it into the coding template, fill {raw_report}, and the assembled prompt asks ChatGPT to extract steps to reproduce, expected vs actual, and the environment. You get a clean ticket body in one copy-paste instead of interrogating the prompt structure from scratch each time a vague report lands in the inbox.

Common pitfalls

Copying the raw template with {placeholders} still in it. Fill every input box first, then copy. The tool shows a live preview so you can confirm no {variable} text leaked into the final prompt before pasting into ChatGPT.
Treating the model label as a hard rule. "Best for Claude" is a hint from our testing, not a requirement. If you only have ChatGPT open, most templates still work fine. The label saves you a comparison test, it does not lock you out.
Leaving an unused placeholder empty, like {context} you do not need. Either delete that line from the assembled prompt or write "none" so the model does not invent context. A literal empty {context} confuses the output more than a clear "no extra context".

Privacy

Everything runs in your browser. The 200+ templates ship as a JavaScript array in the page, and search, category filtering, placeholder extraction, and prompt assembly all happen client-side. Whatever you type into the placeholder inputs stays on your device, is never logged, and is never put into the URL. We do not call OpenAI, Anthropic, or Google. Copy-to-clipboard uses the native browser API.

FAQ

Tool combos

Folks in your role tend to reach for these alongside this tool.

Browse all tools for this role

200+ Prompt Templates for ChatGPT, Claude & Gemini

Blog post outline

What this tool does

Tool details

How to use

1. Input

2. Process

3. Copy / Download

How Prompt Template Library fits into your work

Text jobs

Text checks

Good next steps

Real-world use cases

Stop rewriting the same "summarize this meeting" prompt every Monday

Brief a junior writer who keeps getting the tone wrong

Translate product copy into 5 languages without losing the CTA

Turn a messy bug report into a structured repro for the dev team

Common pitfalls

Privacy

FAQ

Headline Analyzer

Weekly Report Generator

Slogan Generator

Hashtag Generator

Cover Letter Generator

Interview Questions Bank

Resume Template Builder

Invoice Generator

System Prompt Builder

AI Eval Planner

A1Z26 Cipher (Letter ⇄ Number)

Chinese Acupoint Locator