https://github.com/reprompts/repmt
repmt is a lightweight Python library that automatically parses large Python repositories and converts them into clean, structured static prompts โ ideal for AI prompt engineering, documentation, or code intelligence workflows.
https://github.com/reprompts/repmt
beginner-friendly directory open-source parser pip-package prompt prompt-engineering prompts pypi-package pypi-packages python python-library python3 readme-generator repository stream-u streamlit
Last synced: 3 months ago
JSON representation
repmt is a lightweight Python library that automatically parses large Python repositories and converts them into clean, structured static prompts โ ideal for AI prompt engineering, documentation, or code intelligence workflows.
- Host: GitHub
- URL: https://github.com/reprompts/repmt
- Owner: Reprompts
- License: mit
- Created: 2025-04-24T20:24:21.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2025-05-09T08:13:22.000Z (5 months ago)
- Last Synced: 2025-06-07T01:21:36.420Z (4 months ago)
- Topics: beginner-friendly, directory, open-source, parser, pip-package, prompt, prompt-engineering, prompts, pypi-package, pypi-packages, python, python-library, python3, readme-generator, repository, stream-u, streamlit
- Language: Python
- Homepage: https://pypi.org/project/repmt/
- Size: 45.9 KB
- Stars: 3
- Watchers: 1
- Forks: 0
- Open Issues: 4
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
# ๐ Repmt - Repository Prompt Generator for Python Projects

[](https://pypi.org/project/repmt/)
[](https://opensource.org/licenses/MIT)
[](https://pypi.org/project/repmt/)
[](https://github.com/psf/black)**Repmt** is a lightweight tool that analyzes Python repositories and generates **static, structured prompts** โ ideal for technical documentation, AI prompt engineering, onboarding new developers, and building intelligent agents.
Check the Library on Pypi: https://pypi.org/project/repmt/
It provides a **Streamlit-powered UI** to let users select specific files, modules, or directories and generate meaningful, formatted prompts from them.
---
## ๐ Features
- ๐ง **Static Code-to-Prompt Generation** โ Turn codebases into well-structured, static prompts
- ๐ **Streamlit UI** โ Choose what you want to generate: prompt type, modules, files, etc.
- ๐ **Selective Targeting** โ Focus only on selected modules, folders, or classes
- ๐ **Multi-Format Output (Coming Soon)** โ Output to Markdown, HTML, or JSON
- ๐งญ **Prompt Types** โ Tailored for GPT, documentation, architecture summarization, or onboarding
- ๐งผ **Temporary Use Option** โ Auto-cleanup after execution---
## ๐ฆ Installation
```bash
pip install repmt
๐ฅ๏ธ Launch the Streamlit UI
repmtYou will be prompted to:
Select a directory or Python project to analyze
Choose the type of prompt to generate
Choose specific modules, folders, or files to include
Export or copy the generated prompt
โก One-Time Use (Temporary)
You can use repmt temporarily without keeping it installed:pip install repmt && repmt --temp /path/to/repo && pip uninstall repmt -y
The --temp flag auto-uninstalls repmt after use.๐งช Example Use Cases
Feed structured code chunks into LLMs for contextual reasoningDocument legacy or open-source codebases
Generate onboarding prompts for dev teams
Preprocess repos for embedding-based AI agents
๐ License
MIT License โ Free for both open-source and commercial use.
See LICENSE for more details.๐ค Contributing
We welcome contributions!
Please read our Contribution Guidelines before submitting PRs.๐ก๏ธ Security
To report a vulnerability, please email: repromptsquest@gmail.com๐ Acknowledgements
Thanks to the open-source community and prompt engineering ecosystem for inspiring this project.Built with โค๏ธ to make Python repositories more AI-readable and human-friendly.
---
Let me know if youโd like:
- A badge-ready `setup.py` with `console_scripts`
- A `streamlit_app.py` that cleanly links to `repmt`
- Automatic prompt file saving
- Screenshot assets for your GitHub repo visualsHappy open-sourcing!