80 lines
2.8 KiB
Markdown
80 lines
2.8 KiB
Markdown
# Prompter
|
|
|
|
Code smarter with AI—no more messy copy-pasting. Prompter structures your prompts and applies AI changes seamlessly, streamlining your coding workflow.
|
|
|
|
## Why Prompter?
|
|
|
|
- **Too much bloat in your repo?** Stop zipping everything—send only the files that matter.
|
|
- **LLM underperforming?** Cut the noise for sharper, more accurate responses.
|
|
- **Better AI coding?** Select just the right context to optimize results.
|
|
|
|
Prompter empowers you to work efficiently with AI, reducing token waste and improving clarity.
|
|
|
|
## Features
|
|
|
|
- **Advanced File Selection & Token Estimation**
|
|
Precisely filter files and estimate token usage instantly for optimized, cost-effective prompts.
|
|
|
|
- **Optimized XML Prompt**
|
|
Structured file trees, CodeMaps, content, and instructions in XML for maximum LLM clarity.
|
|
|
|
- **Structured XML Diffs**
|
|
Converts LLM-generated XML edits into precise, reviewable diffs—works at any file size.
|
|
|
|
- **Codemap Extraction**
|
|
Scans files locally to extract classes, functions, and references, minimizing tokens and hallucinations. Auto-detects referenced types.
|
|
|
|
- **Mac-Native Performance**
|
|
Built for macOS with native speed and responsiveness—because performance matters.
|
|
|
|
- **Clipboard Integration**
|
|
Copy structured prompts into any AI chat app—your data stays local, no external API needed.
|
|
|
|
- **Works with Any Model**
|
|
Compatible with OpenAI, Anthropic, DeepSeek, Gemini, Azure, OpenRouter, and local models—private and offline when you need it.
|
|
|
|
- **Privacy First**
|
|
Local models, offline scanning, and direct clipboard use—no intermediaries required.
|
|
|
|
## Installation
|
|
|
|
*(Note: Installation steps are assumed based on the VS Code context from other files. Adjust as needed.)*
|
|
1. Clone the repository:
|
|
```bash
|
|
git clone <repository-url>
|
|
```
|
|
2. Open the project in VS Code.
|
|
3. Install dependencies:
|
|
```bash
|
|
npm install
|
|
```
|
|
4. Build the extension:
|
|
```bash
|
|
npm run compile
|
|
```
|
|
5. Press `F5` in VS Code to launch the extension in a development window.
|
|
|
|
## Usage
|
|
|
|
1. Open your project in VS Code.
|
|
2. Use the Prompter interface to select files and estimate tokens.
|
|
3. Generate a structured XML prompt via the clipboard.
|
|
4. Paste into your preferred AI model (e.g., ChatGPT, Claude, or a local LLM).
|
|
5. Apply the returned XML diffs directly through Prompter for seamless integration.
|
|
|
|
## Contributing
|
|
|
|
We welcome contributions! To get started:
|
|
1. Fork the repository.
|
|
2. Create a feature branch: `git checkout -b my-feature`.
|
|
3. Commit your changes: `git commit -m "Add my feature"`.
|
|
4. Push to your branch: `git push origin my-feature`.
|
|
5. Open a pull request.
|
|
|
|
See `vsc-extension-quickstart.md` for development setup and testing details.
|
|
|
|
---
|
|
Built with ❤️ by the Prompter team.
|
|
|
|
Code smarter with AI—no more messy copy-pasting. Prompter structures your prompts and applies AI changes seamlessly, streamlining your coding workflow.
|