A community-curated open-source library of prompts for AI chat models, with a website, CSV/Hugging Face datasets, CLI and MCP server integration.
linshenkx/prompt-optimizer
An AI prompt optimizer that refines prompts, tests and compares their outputs, and saves them as reusable assets, available as web, desktop, Chrome extension, and Docker.
- Stars
- 38,648
- Forks
- 4,393
- Open issues
- 9
- Last commit
- yesterday
What is it?
Prompt Optimizer helps users write better AI prompts and improve the quality of model outputs. Prompts can be written manually, taken from templates, imported from local files, or pulled from Prompt Garden, then optimized in one click with multi-round iteration.
The tool supports both system prompt and user prompt optimization, along with analysis, single-result evaluation, and multi-result compare evaluation to check whether a change actually helps. An advanced testing mode adds context variables, multi-turn conversation tests, and Function Calling for OpenAI and Gemini. Image generation modes cover text-to-image, image-to-image, and multi-image workflows.
The project runs as a pure client-side app that sends requests directly to the configured model provider, and it can also be used as an MCP server for clients such as Claude Desktop. It is available as a website, an Electron-style desktop app with automatic updates for installer builds, a Chrome extension, or a Docker container with optional password protection.
Key features
- 01One-click prompt optimization with multi-round iterative improvement
- 02Supports both system prompt and user prompt optimization
- 03Analysis, single-result evaluation, and multi-result compare evaluation
- 04Connects to OpenAI, Gemini, DeepSeek, Grok, Zhipu AI, SiliconFlow and other models, plus custom OpenAI-compatible APIs
- 05Text-to-image, image-to-image, and multi-image generation
- 06Exposes optimization tools through an MCP server at /mcp
Use cases
- Improving prompts for a small or cheap model so its answers become more specific and structured
- Comparing two prompt versions side by side to confirm which one performs better
- Maintaining a library of tested prompt templates with variables and version history
- Running a self-hosted prompt tool on a private server with password protection
Quick start
- 1
Use the online version (no install needed)
- 2
Run the Docker container with default settings
docker run -d -p 8081:80 --restart unless-stopped --name prompt-optimizer linshen/prompt-optimizer - 3
Install the Chrome extension from the Chrome Web Store
Things to know
- License is AGPL-3.0 per the README badge, though the metadata shows an unrecognized license, so check LICENSE before commercial use.
- Online and browser use stores app data in the browser and sends requests directly to model providers, so you must supply your own working API keys.
- Desktop app is recommended for connecting to local services like Ollama because browser CORS limits apply to web and extension use.
Topics
FAQ
- What is prompt-optimizer?
- An AI prompt optimizer that refines prompts, tests and compares their outputs, and saves them as reusable assets, available as web, desktop, Chrome extension, and Docker. Prompt Optimizer helps users write better AI prompts and improve the quality of model outputs. Prompts can be written manually, taken from templates, imported from local files, or pulled from Prompt Garden, then optimized in one click with multi-round iteration.
- Is prompt-optimizer free to use? What license does it have?
- prompt-optimizer is open source under the Other license. Check the license terms before using it in commercial products.
- Is prompt-optimizer still maintained?
- Actively maintained. The last commit to prompt-optimizer was on Oct 19, 2026.
- What can I use prompt-optimizer for?
- Improving prompts for a small or cheap model so its answers become more specific and structured Comparing two prompt versions side by side to confirm which one performs better Maintaining a library of tested prompt templates with variables and version history Running a self-hosted prompt tool on a private server with password protection
- How do I install prompt-optimizer?
- Use the online version (no install needed) Run the Docker container with default settings: docker run -d -p 8081:80 --restart unless-stopped --name prompt-optimizer linshen/prompt-optimizer Install the Chrome extension from the Chrome Web Store
- Where can I get prompt-optimizer?
- The source code and installation instructions are on GitHub: https://github.com/linshenkx/prompt-optimizer
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A Claude skill that rewrites rough requests into structured, tool-specific prompts for AI tools such as Claude Code, Cursor, Midjourney and ChatGPT.
A public repository of Agent Skills: folders of instructions and resources that Claude loads to perform specialized tasks.
Compresses tool outputs, logs, files and RAG chunks before they reach an LLM, as a library, proxy or MCP server.
A locally installed, power-user front end for chatting with many LLM APIs, image generators and TTS voice models.
An open-source coding-agent CLI that runs in the terminal and works with cloud APIs and local model backends like Ollama.