Prompt House

Prompt House

Let AI pick & use your saved prompts.

23 followers

Tired of messy prompts? Prompt House helps you save, tag, and manage them. Its MCP-enabled feature allows AI to automatically find and use the perfect prompt. Effortless & seamless! Supports clients like Cursor, ChatWise, Cherry Studio, etc.
Prompt House gallery image
Prompt House gallery image
Prompt House gallery image
Free
Launch Team
Anima - OnBrand Vibe Coding
Design-aware AI for modern product teams.
Promoted

What do you think? …

huangyihe
Maker
📌
Hey Product Hunt community! 👋 I'm incredibly excited to officially launch Prompt House today! (https://prompthouse.app) If you're like me, you believe prompts are fundamental to leveraging AI effectively. They're the closest AI usage technique to us, the users! But as you collect more and more valuable prompts – whether you craft them yourself or find them – managing them and, more importantly, using them efficiently becomes a real challenge. Storing them in notes, endlessly searching, and the constant copy-paste dance? It’s clunky and slows you down. I often thought, "There has to be a better way!" That’s exactly why I built Prompt House. I wanted a way to not just store and organize my prompts with titles and tags, but to fundamentally change how I access them by enabling AI to do the heavy lifting. So, what’s the core idea? Why not let your AI models automatically pick the best prompt from your personal library and use it on the fly? With Prompt House, you can: Centralize all your prompts: Save them with clear titles and descriptive tags for refined management. Enable AI-driven retrieval: This is where it gets really cool. Prompt House supports the MCP (Model-Component Protocol). You get a unique link tied to your account. When an AI client needs a prompt: It first fetches your list of prompts (just titles and tags) from the Prompt House server. The AI model then uses these titles and tags to intelligently determine which prompt is most suitable for the current task. Once the choice is made, the model sends another request to the server, specifying which prompt it needs. Finally, the server returns the full content of that selected prompt to the client. It's a beautifully simple flow – typically just two MCP tool calls – making the integration surprisingly straightforward. The goal is to take the friction out of prompt management and invocation. No more manual digging through notes or spreadsheets. Let your AI do the smart work of selecting and calling the right prompt, right when you need it. Prompt House is designed for everyone who uses AI prompts. Whether you're a seasoned AI pro juggling hundreds of prompts, or just starting to explore their power and building up your collection, if you're looking for a smarter, more efficient way to manage and deploy your prompts, I believe Prompt House can help. This is the core functionality I wanted to nail first: solving prompt management and recall in a simple, effective way. I'm thrilled to share Prompt House with you all today and would be very grateful for your thoughts and feedback. How do you currently manage your prompts? What are your biggest challenges? How do you see Prompt House fitting into your AI workflow? Ask me anything! I'll be here all day. Thanks so much for checking out Prompt House!
Emma

@huangyihe 如果作为一个提示词使用和管理的小白用户来说,接触这个工具有几个小小的comments,也可能是没有对这个工具的使用理解到位导致的疑问。

  1. 在prompt house的里面创建的提示词(标题,和标签),和我本地管理的提示词之间有什么关系?两者存在什么关联?

  2. MCP的配置 如何关联使用?

  3. 如果能有一个小小的视频演示对普通使用者可能会更直观。

huangyihe

@emmaca 1、自己存储、管理一大堆提示词也没OK。但是每次使用的时候,需要自己挑选,然后复制粘贴进某个客户端。用Prompt House的话,不需要手动操作,模型会自主选择。2、像ChatWise、Cherry Studio、Cursor等客户端都支持MCP。在Prompt House里获得专属的链接,然后填进客户端的MCP配置,就搞定了。3、我过几天会发布视频详细介绍。