GPT Prompter CLAUDE.md Assistant

A specialized skill for working with GPTPrompter, a PySide6 desktop GUI application that helps assemble sophisticated prompts aligned with OpenAI's GPT-5 Prompting Guide.

What This Skill Does

This skill provides expert guidance on the GPTPrompter codebase - a single-file (~810 lines) Python desktop application for configuring agentic workflows, coding prompts, intelligence controls, metaprompting, few-shot examples, and variable substitution.

Key Capabilities

Architecture Overview

The application has three main components:

1. **Data Layer** (`PromptOptions` dataclass) - Holds all configuration state

2. **Prompt Builder** - Transforms options into final prompt text with variable substitution

3. **GUI Layer** - PySide6 tabbed interface (Basics, Agentic, Coding, Intelligence, Few-shot, Variables & Appendices)

Instructions

When working with the GPTPrompter codebase, follow these guidelines:

1. Environment Setup

Before any code changes:

```bash

Create virtual environment if not exists

python -m venv .venv

source .venv/bin/activate # Windows: .venv\Scripts\activate

Install dependencies

pip install PySide6

Run the application

python gpt5_prompt_designer.py

```

2. Understanding the Code Structure

3. Common Modification Patterns

**Adding a New Prompt Section:**

1. Add field to `PromptOptions` dataclass (lines 48-95)

2. Create `_new_section_block()` method in `PromptBuilder` (lines 100-264)

3. Add method call to `build()` assembly sequence

4. Update `_collect_options()` and `_apply_options()` if GUI controls needed

**Adding a New Tab:**

1. Create `_init_tab_newname()` method following existing tab patterns

2. Add tab creation call in `_init_ui()`

3. Update `_collect_options()` to read new controls

4. Update `_apply_options()` to populate new controls

**Widget Patterns:**

4. Key Features to Understand

**Variable Substitution:**

**Delimiters:**

**Metaprompt Mode:**

**Presets:**

5. Development Workflow

1. Read the relevant section of `gpt5_prompt_designer.py` first

2. Make targeted changes to the specific component (data/builder/GUI)

3. Test manually:

- Launch application

- Configure options across tabs

- Build prompt and verify output

- Test export/import if changed settings

- Verify variable substitution if modified

4. For new features, update all three layers consistently

6. Code Style Guidelines

7. Testing Approach

No automated tests exist. Manual testing checklist:

8. Debugging Tips

Example Usage

**User Request:** "Add a new 'Constraints' section to the prompt"

**Your Response:**

1. Read lines 48-95 to understand `PromptOptions`

2. Add `constraints: str = ""` field to dataclass

3. Add `_constraints_block()` method to `PromptBuilder` (around line 200)

4. Insert method call in `build()` sequence (around line 235)

5. Add UI control in Basics tab (`_init_tab_basics()`)

6. Update `_collect_options()` to read the new control

7. Update `_apply_options()` to populate the control

8. Test by running application and building a prompt with constraints

Important Notes

Constraints