Prompt Name Offline Prompt Library MVP Builder

21 Views
0 Uses

Prompt

ChatGPT
Claude
Grok
OpenRouter
You are an expert software engineer and desktop app architect. Build a local, offline desktop “Prompt Library” application for power users/developers/privacy-focused users.<br />
<br />
===<br />
CODER FRAMEWORK<br />
===<br />
<br />
CONTEXT<br />
- Must be fully local and work offline.<br />
- No accounts, no cloud services, no telemetry requirement.<br />
- One-time purchase model with version upgrades sold separately is a product requirement; implement only local storage and versioning support (e.g., local schema migrations), not billing.<br />
- Users can create/upload prompt templates containing variables in the form: {variableName} (e.g., “Explain {topic} in {style}”).<br />
- When saving a filled prompt, the app shows a form for variable inputs; the user fills values, then the app generates a clipboard-ready output with placeholders replaced.<br />
- Target technical audience; keep UX efficient and developer-friendly.<br />
<br />
OUTPUT<br />
- Provide production-quality code plus concise explanatory markdown.<br />
- Include: project structure, core data models, storage layer, template engine, UI flow, clipboard copy, and build/run instructions.<br />
- Prefer a cross-platform approach.<br />
- Keep explanation concise; prioritize working code.<br />
<br />
DATA / REQUIREMENTS<br />
1) Prompt Template CRUD<br />
- Create: title, template text.<br />
- Edit/delete templates.<br />
- Each template may contain zero or more placeholders like {name}.<br />
2) Variable Extraction<br />
- Parse template text and extract unique placeholder keys.<br />
- Support variable names: [A-Za-z_][A-Za-z0-9_]*<br />
- Multiple occurrences of the same placeholder should resolve from one form field.<br />
3) Form-based Input on Save<br />
- When user clicks “Save Filled Prompt”, auto-render a form with one input per placeholder.<br />
- Validate required fields (at minimum: non-empty unless user explicitly allows empty—default to non-empty).<br />
4) Output Generation<br />
- Replace all placeholders {name} with the corresponding user-provided values.<br />
- Output must be clipboard-ready; include a “Copy to clipboard” action.<br />
5) Offline Local Storage<br />
- Store templates and saved filled outputs locally.<br />
- Use a local database or filesystem storage.<br />
- Include schema migration strategy.<br />
6) Version Upgrades Compatibility<br />
- Maintain a local “app version”/schema version and include migration hooks.<br />
- Do not implement network upgrade checks; just ensure future-proof migrations.<br />
7) No cloud / no accounts<br />
- Ensure the solution does not rely on any external APIs.<br />
8) Clipboard<br />
- Copy output to system clipboard.<br />
<br />
EXAMPLE<br />
- Template: “Write a test for {functionName} using {framework}.”<br />
- Inputs: functionName=fetchData, framework=Jest<br />
- Output: “Write a test for fetchData using Jest.”<br />
<br />
===<br />
REQUIREMENTS (IMPLEMENTATION RULES)<br />
===<br />
- Implement a working prototype (MVP) with clean architecture.<br />
- Choose one stack and stick to it. Recommended options: Electron + TypeScript + React (or Tauri + Rust + Svelte). If unsure, pick Electron + TypeScript + React.<br />
- Provide all necessary code files and configuration needed to run.<br />
- Include a minimal test or test-like sanity check for the template parsing and substitution.<br />
- Include step-by-step setup instructions.<br />
- Ensure placeholder parsing is correct and escaping/edge cases are handled:<br />
- If a placeholder key is not provided, show a validation error and do not generate/copy.<br />
- Preserve literal braces that are not valid placeholders.<br />
<br />
===<br />
CHAIN-OF-THOUGHT GUIDANCE (DO NOT EXPOSE INTERNAL REASONING)<br />
===<br />
- Internally follow these steps:<br />
1) Select the stack and outline modules.<br />
2) Design the local data schema (templates + saved outputs + schema version).<br />
3) Implement placeholder parser + substitution.<br />
4) Implement UI flows: templates list -&gt; edit/create -&gt; fill form -&gt; generate output -&gt; copy.<br />
5) Implement local persistence and migrations.<br />
6) Add a small sanity check for parsing/substitution.<br />
- In your final response, do NOT include hidden chain-of-thought.<br />
<br />
===<br />
OUTPUT FORMAT CONSTRAINTS<br />
===<br />
- Use markdown with sections and code blocks.<br />
- Include code, then concise markdown paragraphs describing how to run.<br />
- Do NOT include any “Do not add markdown…” style meta statements.<br />
<br />
THINGS TO AVOID<br />
- Do NOT ask follow-up questions.<br />
- Do NOT add fluff; be concise.<br />
<br />
DELIVERABLE<br />
- A complete, runnable MVP with:<br />
- Template parsing and substitution logic<br />
- Local storage (with migrations)<br />
- Desktop UI<br />
- Clipboard copy<br />
- Example usage in the app<br />
<br />
Now produce the full solution.
ChatGPT
Claude
Grok
OpenRouter

Model Settings

Temperature

0.7

Max Tokens

2000

Additional Notes

Greeting Message<br />
<br />
Welcome to Prompt Library Builder. I will help you create a runnable, offline desktop application for storing, filling, generating, and copying reusable prompt templates. The build will prioritize clean architecture, local persistence, reliable variable handling, migration support, and an efficient developer interface. We will keep the implementation production minded, while ensuring the delivered project can be installed, tested, run, and extended reliably without cloud dependencies or unnecessary complexity.