PROMPT++

PROMPT++

Refines the prompts you write for AI models, using a metaprompt router and a refiner that call open models such as Zephyr and Llama. For anyone who wants clearer prompts.

Made by baconnier on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: PROMPT++ helps you improve the prompts you give to AI models. The repository has a metaprompt router, a set of sample metaprompts and a prompt refiner, and the Space names several open chat models such as Zephyr and Llama. The README is a single line, so the controls are not described there.

Who it suits: It suits people who write prompts often and want a second pass, such as writers, marketers and developers. It also suits prompt engineers who want to see how a metaprompt-based refiner is built.

Industry | Who uses it | How they use it | Example
Writing | Content writers | Tighten a rough prompt before using it | Paste a draft prompt and read the refined version
Marketing | Marketers | Make prompts for campaign copy clearer | Refine a prompt that asks for ad text
Software | Developers | Improve prompts that go into an application | Test a few versions of one instruction
Research | Prompt engineers | Study a metaprompt router and refiner | Read the router and refiner files

Anyone who already has a tested prompt library can skip it, since this is a first-pass refiner.

Where to use it: The app runs in two places.

• A web browser, on the page of the Space at Hugging Face
• Windows, macOS or Linux, as a local copy run with Python

Use case | Who | What they get back
Prompt refinement | Writers | A clearer version of a prompt you wrote
Model choice | Developers | A way to try refinement with different open models
Code study | Prompt engineers | A router, sample metaprompts and a refiner to read

How to set it up: The online page needs nothing installed. A local copy needs the following.

• Python and pip
• Windows, macOS or Linux
• The packages listed in requirements.txt
• Access to the models the Space names, which the source does not describe

1. Install Python.
2. Install the Hugging Face hub package and download a copy of the Space.
3. Move into the new folder and install Gradio 4.44.1 and the packages in requirements.txt.
4. Start the app with python app.py.

pip install -U huggingface_hub
hf download baconnier/prompt-plus-plus --repo-type space --local-dir prompt-plus-plus
cd prompt-plus-plus
pip install gradio==4.44.1
pip install -r requirements.txt
python app.py

How to use it: Open the page, paste the prompt you want to improve, pick a model if the page offers one, and run it. The source does not describe the controls. On a local copy, this command starts the page.

python app.py

Tool | What it does
prompt_refiner.py | The file that refines a prompt
metaprompt_router.py | The file that picks a metaprompt for the task
metaprompt_sample.py | Sample metaprompts the app can use
app.py | Holds the Gradio page

Example: You paste a short, vague prompt asking for a product description. The app returns a rewritten prompt that states the task more clearly, which you can copy into your own AI tool.

For newcomers and experts: If you are new to this, start with a prompt you already use and see how the refined version differs. Keep what helps and discard what does not.

If you know the field, the repository splits the work into a router that selects a metaprompt, sample metaprompts and a refiner that applies one. The Space names models from the Zephyr and Llama families. The Space runs on basic CPU hardware, so the named models are presumably reached through a hosted service, which the source does not state.

Paste a rough prompt
The router picks a metaprompt
The refiner rewrites the prompt
You copy the improved version

Good to know: The Space page gives one line of description. Read the code for settings, and check the model terms before you rely on the output.

• License: Apache 2
• Cost to run: none named
• Data it reaches: the prompts you type, which are passed to a model the Space names
• Hardware: basic CPU at Hugging Face

On Hugging Face, 13 November 2026

• Likes: 247
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/baconnier/prompt-plus-plus
• Files: https://huggingface.co/spaces/baconnier/prompt-plus-plus/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.