linshenkx /prompt-optimizer

An AI prompt optimizer for writing better prompts and getting better AI results.

★ 41,636⑂ 4,675👁 129⑃ 113● 238▤ 62
TypeScriptNOASSERTIONCreated 2025-02-12prompt.always200.com
Download: v2.15.2v2.15.2 ↓2.15.22026-11-09
+93
Daily
+656
Weekly
41,636
stars
4,675
forks
129
watchers
113
prs
238
issues
62
releases

README

Prompt Optimizer 🚀



English | 中文


!Chrome Web Store Users





!GitHub forks


Website | Online Optimizer | Prompt Garden | Docs | Quick Start | Chrome Extension | 💖 Support

First-time Tutorial | Development Docs | Vercel Deployment Guide | Cloudflare Deployment Guide | MCP Deployment Guide | DeepWiki Docs | ZRead Docs



 📖 Project Introduction

Prompt Optimizer is a powerful AI prompt optimization tool that helps you write better AI prompts and improve the quality of AI outputs. It supports four usage methods: web application, desktop application, Chrome extension, and Docker deployment.

Prompts can start from manual writing, templates, local imports, or sources such as Prompt Garden. Prompt Optimizer is where those prompts are optimized, tested, evaluated, and saved as reusable prompt assets.

 🎥 Feature Demonstration


  1. Hard-Nosed Reviewer: Turn Agreement into Useful Critique
  Starting from a minimal English role prompt, optimization pushes a small model away from generic pushback and toward a clearer, more structured review that surfaces weak assumptions, evidence gaps, and concrete revision advice.
  
  
  2. Marketplace Bargaining Reply: Let Variables Change the Strategy
  With a single reusable prompt template, you can swap in item details, price anchors, buyer offers, tone, and negotiation goals for different marketplace conversations. After optimization, the same small model does a better job turning those variables into a clearer, more transaction-ready reply instead of a generic helper-style response.
  
  
  3. Text-to-Image: Optimize a One-Line Idea into a More Directable Key Visual Prompt
  This is not just prompt expansion. Starting from a vague one-line idea, Prompt Optimizer adds clearer subject cues, spatial relationships, and mood anchors. The left side is simply “a floating library in the night sky,” while the optimized version gives the model a more directed fantasy composition that feels closer to a reusable key visual than a lucky generic image.
  


 ✨ Core Features

- 🎯 Intelligent Optimization: One-click prompt optimization with multi-round iterative improvements to enhance AI response accuracy
- 📝 Dual Mode Optimization: Support for both system prompt optimization and user prompt optimization to meet different usage scenarios
- 🔄 Analysis and Compare Evaluation: Supports analysis, single-result evaluation, and multi-result compare evaluation to help determine whether a prompt has truly improved
- 🤖 Multi-model Integration: Support for mainstream AI models including OpenAI, Gemini, DeepSeek, Grok, Zhipu AI, SiliconFlow, MiniMax, etc.
- 🖼️ Image Generation: Support for Text-to-Image (T2I), Image-to-Image (I2I), and Multi-Image generation with models like Gemini, Seedream, Grok
- 🌱 Prompt Sources: Start from manual writing, templates, local imports, or Prompt Garden import codes
- ⭐ Smart Favorites: Resource-aware prompt assets with version history, reproducible examples, media support, source binding, and works

Top issues

Discussions

Latest release

v2.15.2 v2.15.2 2026-11-09
English

 Summary

This patch restores automatic wrapping and Stop controls for streaming output, updates two security-sensitive dependencies, and includes Garden browsing improvements and optional GPT Image 2.6 GIF output added since v2.15.1.

 Highlights

- Read long streamed responses and source views without horizontal overflow.
- Stop an individual text stream in ordinary workspaces and fullscreen views.
- Search and browse Garden inspiration and Arena evaluation prompts in place.
- Select GIF output for compatible GPT Image 2.6 endpoints while retaining PNG as the default.

 Product Upda

Snapshot: 2026-11-16T09:23:14Z · Data license Apache-2.0 · JSON