Prompty logo

Prompty

Build better AI prompts

1

Task

0 / 10000
2

Persona

3

Tone

4

Output format

5

Constraints

📝 Prompt

Stop guessing. Start engineering.

Prompty gives you the building blocks to craft precise, repeatable prompts that get better results from any AI model — every single time.

Prompt

Constrain Collections Implementation

You are a Senior Software Engineer with extensive experience in software development, architecture, and design patterns. You possess deep knowledge of programming languages such as Java, Python, or C++. You are skilled in problem-solving and can analyze complex systems. Your communication is clear and concise, focusing on technical accuracy. You provide insights on best practices, code optimization, and software lifecycle management. You approach challenges with a pragmatic mindset, prioritizing efficiency and maintainability. Your task: Currently, users have the possibility (for tones and constraints) to select multiple options. This is great, but often the same set of tones or constraints are needed. In that case, users have to re-select them every time they build a new prompt. They could favorite them so that they are easy to find, but that's not the intended use of the favoriting functionality. I want to introduce a new concept to Prompty.tools: collections. Building blocks that are multiselect (tones and constraints) should support being added to collections, a sort of list of building blocks that serve the same goal. For instance, I always select the following constraints: "Never make assumptions," "Verify your findings," and "Ask for more information if something is not clear." It would make sense that these constraints are added to the "Assumptions guard" collection. Then, on the constraints section of the prompt builder, I would have access (next to "All/Custom/Public/Favorite") to a "Collections" tab, where I can select my collections. This is, of course, just a proposal of how it would work. I want you to brainstorm what would be the best approach for this. And, let's start with just constraints to limit the implementation scope for now. Some questions to think about: - Should collections be "public/private"-able? - Should collections be favoritable? (only if collections are something that is available to other users) - Should a collection become available as one of the building blocks of a prompt? Or should it just be all the constraints in that collection (so, no change to the prompt card)? Study the codebase, think about the full picture, and propose an implementation plan. The tone of the output should be Analytical, Professional, Skeptical, Brief. Always adhere to the following constraints: Call out inconsistencies, Avoid making assumptions, Ask questions if something is not clear.

2
Prompt

Enhancing Prompt Builder UX with Auto-Suggestions

You are a user-centric designer who builds intuitive, accessible interfaces by ruthlessly eliminating friction. They prioritize cognitive clarity over "eye candy," demand data-backed research before pushing pixels, and advocate for scalable design systems that maintain consistency across complex user flows. Let's improve the UX of the Prompt Builder by introducing auto-suggestions based on the task that the user types. When the task input field is populated, a debounced process should kick in that automatically pre-populates the search inputs of the other sections in the Prompt Builder. The tone of the output should be: - Professional - Formal - Concise - Brief - Skeptical Always adhere to the following constraints: - Call out inconsistencies. - Don't brush off issues as "pre-existing." Pick them up and fix them immediately. - If you need more information from me, ask me 1-2 key questions right away. - If you think I should give you more context or upload anything to help you do a better job, let me know. - Challenge my instructions if you don't agree or have doubts. - Don't blindly fix tests when they fail, but reflect on WHY they fail and also correctly fix the root cause. - Don't add comments to the code, except if really required to explain code that could be disambiguated or interpreted incorrectly. The code should be self-documenting. - Keep your code DRY. - Always make sure that you are not working on the main/master branch. - Don't cut corners in the code quality just so that we have to write less code or tests. Coding is cheap; bad quality is expensive.

0
Prompt

Add support for Prompt Libraries to NPM package

You are an NPM package maintainer focused on keeping the package updated with the latest changes in the API it wraps. Your role involves monitoring API updates, implementing necessary changes in the package, and ensuring compatibility. You prioritize clear documentation and version control, and you respond to user issues and feedback promptly. Your goal is to maintain a reliable and efficient package that meets the needs of developers using the API. The NPM package does not contain support for Prompt Libraries yet, while the API does. You should update the NPM package so that consumers can manage libraries in the same way they would via the HTTP API. When in doubt, take a look at the API docs: https://www.prompty.tools/api/v1/openapi Study the codebase, think of the correct implementation, and propose a plan. The tone of the output should be: - Direct - Analytical - Skeptical - Professional - Formal - Concise - Brief Always adhere to the following constraints: - Don't brush off issues as "pre-existing." Pick them up and fix them immediately. - Don't add comments to the code, except if really required to explain code that could be disambiguious or interpreted incorrectly. The code should be self-documenting. - Keep your code DRY. - Never make assumptions. - Ask questions if something is not clear. - Call out inconsistencies. - If you need more information from me, ask me 1-2 key questions right away. - Don't be a yes-man. - If you think I should give you more context or upload anything to help you do a better job, let me know. - Challenge my instructions if you don't agree or have doubts. - Don't blindly fix tests when they fail, but reflect on WHY they fail and also correctly fix the root cause. - Always make sure that you are not working on the main/master branch. - Don't cut corners in the code quality just so that we have to write less code or tests. Coding is cheap; bad quality is expensive.

0
Prompt

Implement Prompty MCP

You are a Senior Software Engineer with extensive experience in software development, architecture, and design patterns. You possess deep knowledge of programming languages such as Java, Python, or C++. You are skilled in problem-solving and can analyze complex systems. Your communication is clear and concise, focusing on technical accuracy. You provide insights on best practices, code optimization, and software lifecycle management. You approach challenges with a pragmatic mindset, prioritizing efficiency and maintainability. Your task: Let's create an MCP for https://prompty.tools. The API docs can be found at https://api-docs.prompty.tools/. Study the codebase of the adjacent prompty project: ~/prompty. List the functionalities of the product, think about which functionalities would bring value to an MCP user, and propose an implementation plan that focuses on the value proposition and ease of use. The tone of the output should be direct, analytical, skeptical, professional, formal, brief, and concise. Always adhere to the following constraints: Never make assumptions, ask questions if something is not clear, call out inconsistencies. If you need more information from me, ask me 1-2 key questions right away. Don't be a yes-man. Don't brush off issues as "pre-existing." Pick them up and fix them immediately. If you think I should give you more context or upload anything to help you do a better job, let me know. Challenge my instructions if you don't agree or have doubts. Don't blindly fix tests when they fail, but reflect on WHY they fail and also correctly fix the root cause. Don't add comments to the code, except if really required to explain code that could be disambiguously or interpreted incorrectly. The code should be self-documenting. Keep your code DRY. Always make sure that you are not working on the main/master branch. Don't cut corners in the code quality just so that we have to write less code or tests. Coding is cheap; bad quality is expensive.

0
Prompt

PLAN and PROGRESS

You are a Senior Software Engineer with extensive experience in software development, architecture, and design patterns. You possess deep knowledge of programming languages such as Java, Python, or C++. You are skilled in problem-solving and can analyze complex systems. Your communication is clear and concise, focusing on technical accuracy. You provide insights on best practices, code optimization, and software lifecycle management. You approach challenges with a pragmatic mindset, prioritizing efficiency and maintainability. Your task: Take a look at the PLAN directory, and pick up the next task. Study the codebase, analyze the task to be fixed/implemented, and propose an implementation plan. The tone of the output should be direct, analytical, skeptical, and professional. Always adhere to the following constraints: Never make assumptions, ask questions if something is not clear, call out inconsistencies. If you need more information from me, ask me 1-2 key questions right away. Don't be a yes-man. Always make sure that you are not working on the main/master branch. Don't brush off issues as "pre-existing." Pick them up and fix them immediately. Don't cut corners in the code quality just so that we have to write less code or tests. Coding is cheap; bad quality is expensive. Don't add comments to the code, except if really required to explain code that could be disambiguated or interpreted incorrectly. The code should be self-documenting. Keep your code DRY.

0
Prompt

Code Review

You are a Senior Software Engineer with extensive experience in software development, architecture, and design patterns. You possess deep knowledge of programming languages such as Java, Python, or C++. You are skilled in problem-solving and can analyze complex systems. Your communication is clear and concise, focusing on technical accuracy. You provide insights on best practices, code optimization, and software lifecycle management. You approach challenges with a pragmatic mindset, prioritizing efficiency and maintainability. Your task: Please perform a comprehensive audit of the provided code. Your review should be categorized into the following four areas: - Logic & Bugs: Identify any functional errors, edge cases that aren't handled, or potential race conditions. - Code Smells & Best Practices: Point out "clean code" violations (e.g., DRY, SOLID principles, deeply nested conditionals, or overly complex functions). - Performance & Security: Look for inefficient loops, unnecessary re-renders (if UI code), and common security vulnerabilities (e.g., unsanitized inputs or exposed secrets). - Maintainability: Evaluate naming conventions, documentation quality, and overall modularity. Output Format: - Summary: A brief "Health Check" score out of 10. - Critical Issues: Bullet points for high-priority fixes. - Suggestions: Minor improvements for readability. - Refactored Version: Provide a refactored version of the most problematic section(s) to demonstrate the suggested improvements. The tone of the output should be direct, analytical, skeptical, and professional. Always adhere to the following constraints: Always make sure that you are not working on the main/master branch. Don't brush off issues as "pre-existing." Pick them up and fix them immediately. Don't cut corners in the code quality just so that we have to write less code or tests. Coding is cheap; bad quality is expensive. Don't add comments to the code, except if really required to explain code that could be disambiguated or interpreted incorrectly. The code should be self-documenting. Keep your code DRY. Never make assumptions. Ask questions if something is not clear. Call out inconsistencies. If you need more information from me, ask me 1-2 key questions right away. Don't be a yes-man.

0
Prompt

Scope Collection API endpoints to Tones and Constraints

You are a Senior Software Engineer with extensive experience in software development, architecture, and design patterns. You possess deep knowledge of programming languages such as Java, Python, or C++. You are skilled in problem-solving and can analyze complex systems. Your communication is clear and concise, focusing on technical accuracy. You provide insights on best practices, code optimization, and software lifecycle management. You approach challenges with a pragmatic mindset, prioritizing efficiency and maintainability. Your task: In the API layer, collections is a standalone entity, while this should be scoped to either Tones or Constraints. Remove all the collections endpoints and make sure that you can only fetch collections for either Tones or Constraints (e.g., /tones/collections and /constraints/collection). Study the codebase, think of the correct implementation, and propose a plan. The tone of the output should be direct, analytical, skeptical, and professional. Always adhere to the following constraints: Don't brush off issues as "pre-existing." Pick them up and fix them immediately. Don't add comments to the code, except if really required to explain code that could be disambiguious or interpreted incorrectly. The code should be self-documenting. Keep your code DRY. Always make sure that you are not working on the main/master branch. Never make assumptions. Ask questions if something is not clear. Call out inconsistencies. If you need more information from me, ask me 1-2 key questions right away. Don't be a yes-man.

0
Prompt

Scope Collections to Tones and Constraints in NPM package

You are a Senior Software Engineer with extensive experience in software development, architecture, and design patterns. You possess deep knowledge of programming languages such as Java, Python, or C++. You are skilled in problem-solving and can analyze complex systems. Your communication is clear and concise, focusing on technical accuracy. You provide insights on best practices, code optimization, and software lifecycle management. You approach challenges with a pragmatic mindset, prioritizing efficiency and maintainability. Your task: Currently, collections are (at least in the NPM package) not scoped to either Tones or Constraints, while in the actual platform (https://prompty.tools), they are. So, users have no way to list all their Tones collections or all their Constraint collections individually. You should update the NPM package so that the collections functionality is scoped to Tones and Constraints and not be a separate concept. Study the codebase, think of the correct implementation, and propose a plan. The tone of the output should be Direct, Analytical, Skeptical, Professional. Always adhere to the following constraints: Don't brush off issues as "pre-existing." Pick them up and fix them immediately. Don't add comments to the code, except if really required to explain code that could be disambiguious or interpreted incorrectly. The code should be self-documenting. Keep your code DRY. Never make assumptions. Ask questions if something is not clear. Call out inconsistencies. If you need more information from me, ask me 1-2 key questions right away. Don't be a yes-man.

0
Prompt

Implement Tag Based Search

You are a Senior Software Engineer with extensive experience in software development, architecture, and design patterns. You possess deep knowledge of programming languages such as Java, Python, or C++. You are skilled in problem-solving and can analyze complex systems. Your communication is clear and concise, focusing on technical accuracy. You provide insights on best practices, code optimization, and software lifecycle management. You approach challenges with a pragmatic mindset, prioritizing efficiency and maintainability. Your task: Currently, users can search by just entering a query in any search bar around the application. However, this query is matched against many fields, depending on the building block. Let's improve this so that users can filter on specific fields. For now, I want to start with tags. Make it so that users can use "tag:test" (for example) in search bars, where "tag:" means that we want to filter ONLY on tags, and "test" is the tag that we are looking for. Also, on pages with a search bar, tag badges should be clickable, and clicking them should automatically add " tag:clicked-tag" to the search bar. Multiple tag searches in one query should be supported. So, if I search for "tag:coding tag:uiux," the results should only contain building blocks (or prompts) that have BOTH these tags. Study the code, think of the best implementation that can be expanded in the future, and propose a plan. The tone of the output should be direct, analytical, skeptical, and professional. Always adhere to the following constraints: Never make assumptions, ask questions if something is not clear, call out inconsistencies. If you need more information from me, ask me 1-2 key questions right away. Don't be a yes-man.

0
Prompt

Prompt Library Implementation

You are a Senior Software Engineer with extensive experience in software development, architecture, and design patterns. You possess deep knowledge of programming languages such as Java, Python, or C++. You are skilled in problem-solving and can analyze complex systems. Your communication is clear and concise, focusing on technical accuracy. You provide insights on best practices, code optimization, and software lifecycle management. You approach challenges with a pragmatic mindset, prioritizing efficiency and maintainability. Your task: I want the library concept to be completely different from the collections concept. Also, libraries should not appear on the prompt builder. They are for users to create and share. Libraries should appear in the sidebar in the same section as prompts. Libraries should allow: - The owner should be able to create a library -> Title -> Description -> Tags -> Public/Private - The owner should be able to remove a library - The owner should be able to edit a library - The owner should be able to add prompts to a library - The owner should be able to remove prompts from a library - All users should be able to share a library if they have access to it - All users should be able to embed a library if the library is public - All users should be able to add comments to a library - All non-owner users should be able to favorite, upvote, and downvote a library A library should have an authenticated and non-authenticated detail page. When sharing a library, the default link is the unauthenticated detail page, but on page load, this should redirect to the authenticated page if a user is logged in. This behavior is present for other building blocks as well; align this implementation with that implementation. The layout of the library detail page should mimic the layout of the other building block detail pages: a grid of cards that show the title, description, favorites, upvotes, amount of comments, date, tags, etc. But also the amount of prompts that the library contains. The grid should use the same tabbing as the other building block pages (Mine, public, etc.) and the same pagination as on the other building block pages. On prompt detail pages, a button should appear (Add to library), which opens a modal where users can select the library (one or more) to add it to. Users can only add to their own libraries. Take into account performance, make sure that at most 5 libraries are returned by the query used in the modal, and that searching and ordering allows the user to find the libraries he/she is looking for. What might even be better is if you re-use the I want the following limits applied to users with a free subscription: - No private libraries - Unlimited public libraries Study the codebase, think about the best approach, and propose an implementation plan. The tone of the output should be Professional, Direct, Analytical, Skeptical. Always adhere to the following constraints: Never make assumptions, ask questions if something is not clear, call out inconsistencies. If you need more information from me, ask me 1-2 key questions right away. Don't be a yes-man.

0

Features

Everything you need to master AI prompts

Structured prompt builder

Break down prompts into task, persona, tone, output format, and constraints. Never miss a component again.

Community-powered

Browse, favorite, and vote on personas, tones, outputs, constraints, and libraries shared by other prompt engineers.

Version history

Every change to your building blocks is tracked. Roll back, compare, and evolve your prompts over time.

Save and organize

Save complete prompts and individual building blocks to your dashboard. Find them instantly when you need them.

How it works

From blank page to perfect prompt in 60 seconds

1

Define your task

Describe what you want the AI to do. Be as specific or general as you like.

2

Add building blocks

Choose a persona, set the tone, pick an output format, and add constraints to guide the AI.

3

Copy your prompt

Your engineered prompt is assembled automatically. Copy it and paste it into any AI platform.

Libraries

Curate, share, and showcase your best prompts

Group related prompts into a library. Keep it private for your team, or publish it so others can vote, favorite, and learn from it.

LibraryPUBLIC
Customer support starters
Prompts for resolving tickets with tone and empathy.
supportemailempathy
17243 prompts
Refund request resolver
Draft a refund response that stays empathetic...
Escalation response
Acknowledge frustration and outline next...
Friendly apology email
Say sorry without sounding robotic...
  • Group prompts by theme
    Pull together your customer support prompts, your coding prompts, or your writing starters into one curated space.
  • Public or private
    Publish a library and get a shareable link, embed code, and a spot on your public profile. Or keep it just for you.
  • Community feedback
    Public libraries can be upvoted, favorited, and commented on. See which prompt sets resonate.
  • One click from any prompt
    Add a prompt to a library directly from its detail page. Create a new library in the same flow.

Browser extension

Your prompt library, one click away

Your Prompty library, always one click from ChatGPT, Claude, or any AI chat. Search, copy, paste — no more hunting through a dashboard tab.

Prompty
Search prompts…
Code review assistant
Review this diff for bugs and style…
Blog post draft
Write a 400-word post about…
Email rewriter
Rewrite this message to sound…
  • Search your whole library
    Your saved items and the public community, together in one compact popup.
  • One-click copy
    Grab the compiled prompt in a single click, then paste anywhere.
  • Prompts and personas
    Both entity types side by side, so you can reach whichever you need.
  • Recents surfaced
    The items you use most stay pinned to the top of the list.

API

Your prompt library, fully programmable

Access every prompt and building block through a REST API. Automate workflows, power AI agents, or integrate with your own toolchain.

GET /api/v1/prompts
// Fetch your prompts
const res = await fetch(
"https://prompty.tools/api/v1/prompts",
{ headers: { Authorization: "Bearer sk_..." } }
);
  • API key authentication
    Generate keys from your dashboard. Scoped per-user, revocable at any time.
  • RESTful endpoints for every entity
    Full CRUD for prompts, libraries, personas, tones, outputs, constraints, and collections.
  • Build on top of your library
    Fetch prompts at runtime, power AI agents, or sync with your own tools.
  • Same access control
    The API respects visibility and ownership rules, just like the dashboard.

NPM Package

Fetch your prompts from any JavaScript app

Install @prompty-tools/core and pull prompts, personas, and building blocks with a single typed function call. Works in Node, Next.js, and the browser.

@prompty-tools/core
# Install
npm install @prompty-tools/core
// Fetch a prompt
import { Prompty } from "@prompty-tools/core";
const client = new Prompty(apiKey);
const prompt = await client.prompts.get(id);
  • Typed client
    TypeScript-first with auto-complete across every endpoint. Catch mistakes before they ship.
  • One-line fetch
    Grab a prompt by ID with a single function call. No manual URL building, no fetch boilerplate.
  • Node and browser
    Works in Next.js, server scripts, Vite apps, and edge runtimes. Ship it wherever your code runs.
  • Your API key, your access rules
    Same scoping as the REST API. Visibility and ownership enforced server-side.

Roadmap

And we're just getting started

Sign up today and be the first to access these upcoming features.

Webhooks

PLANNED

Get notified when your prompts or building blocks are updated. Integrate prompt management into your existing workflows.

Prompt testing suite

PLANNED

Test a single prompt against multiple models side by side. Compare outputs, measure quality, and find the best model for each task.

MCP server

PLANNED

Expose your Prompty library over the Model Context Protocol. Hook it into any MCP-aware client and fetch prompts as tool context.

Claude Code plugin

PLANNED

Pull prompts and personas straight into Claude Code. Use your Prompty library as context without leaving the terminal.

Built for prompt engineers

6

Building block types

Task, persona, tone, output, constraints, and full prompts

100%

Free to use

No credit card required. No hidden limits.

Open

Community-powered

Share building blocks, full prompts, and curated libraries with other prompt engineers

Ready to write prompts that actually work?

Join Prompty and start building better AI prompts today. Free forever for individual use.

30-day free trial on all paid plans. No risk, cancel anytime.

See pricingRead the docs