Rebuild chatbot prompt

You are the greatest software engineer in the world. Your job today is to write a prompt for CLIne, an agentic AI coding assistant that will be used to rebuild the project that is going to be redone based on the attached repo. The attached repo has a working prototype and should be studied to see what it did correctly and what it did incorrectly and those considerations should then be used to inform a new plan of action which will inform the context you will provide in the prompt you are going to give CLIne. Some things about the repo, it is a chatbot which allows you to use a system prompt and save them and use change them out, it uses Ollama to allow you to choose any loaded local model you want. It has many features for the UI and they should be analyzed and rearranged for better UI/UX. The TTS option can be altered to be a better option than edge-tts. I want this prompt to have instructions for CLIne to compose the following documents in a folder in the root of the project: ai_guidelines.md - which describes the guidelines for the AI to follow when it is coding, there should be a checklist.md where CLIne can keep track of its progress, there should be an architecture.md where all the details of the application are first planned out. The project should incorporate as many features as you can feasibly or realistically expect it to be able to create. prompts.md will hold a series of prompts for CLIne to perform, testing.md will have instructions on how to do proper testing for this application fully, security.md will address all security concerns, design.md will outline and describe all of the design ideas for frontend to be implemented and how thing should be arranged and structured, there should be a uxui.md which outlines the uxui requirements, there should be a readme.md which is top notch for github with all the modern standards as well as any other planning documents that would help the process. --- Now take this and formulate a prompt I will give to an LLM in order to formulate the prompt which I will give to another LLM to formulate the prompt for CLIne

Here’s a **structured meta-prompt** you can feed into an LLM, so that **it will generate the final prompt you’ll give to another LLM, which will then generate the actual CLIne prompt** for building the new project. --- ## ✅ Meta-Prompt (What You Give to the LLM Right Now) **You are designing a prompt that will be given to another LLM in order to generate the actual CLIne prompt for rebuilding a chatbot project.** Your job is to output only the **final prompt text** that will be used for that second LLM — do not execute it, do not explain it, only write it. The resulting prompt you generate must instruct the second LLM to: --- ### 🎯 **1. Analyze the Existing Repository** - Study the attached repository (chatbot using Ollama, system-prompt management, UI features, TTS using edge-tts). - Identify: - What was done correctly (architecture, modularity, UI ideas, TTS handling). - What was done poorly (UI/UX flaws, code structure, scalability issues, missing error handling, inefficient API usage). - Use this analysis to form a **new architecture plan and development strategy** for CLIne. --- ### 🏗️ **2. Instruct CLIne to Rebuild the Project With Major Improvements** The prompt generated must tell CLIne to: - Create a fully refactored project using modern coding standards. - Improve the chatbot system by: - Better system-prompt management and saving. - Using Ollama with selectable models. - Improved TTS (not edge-tts; suggest better alternatives like Piper, Coqui TTS, etc.). - Stronger error handling and async handling. - Redesign the frontend with improved UI/UX, accessibility, smooth state management, and animations where appropriate. --- ### 📂 **3. CLIne Must Generate the Following Documents in `/docs` or root folder:** | File | Purpose | |------|---------| | `ai_guidelines.md` | Rules for how the AI agent should write, refactor, and review code. | | `checklist.md` | A living progress tracker for CLIne. | | `architecture.md` | High-level system architecture, data flows, components, backend/frontend layers. | | `prompts.md` | Reusable task prompts for CLIne (coding, debugging, testing, deployment, etc). | | `testing.md` | Full test architecture: unit, integration, UI testing, mock data, LLM testing. | | `security.md` | Security concerns: prompt injection, API abuse, XSS, CORS, user data risks. | | `design.md` | Visual design system, component structure, layout decisions, color/font hierarchy. | | `uxui.md` | User experience requirements: navigation flow, modal/dialog patterns, prompt workflows. | | `readme.md` | Professional open-source-level documentation (installation, features, screenshots, demo, contribution). | | *(Optional)* `deployment.md` | CI/CD, Docker, hosting (Render, Vercel, Netlify), environment variables. | | *(Optional)* `roadmap.md` | Future release planning and feature milestones. | --- ### ⚙️ **4. Development Expectations for CLIne** The instructions inside the prompt must require CLIne to: - Use modular, scalable architecture. - Use TypeScript on frontend/React, Python or Node for backend (unless otherwise decided). - Implement state management (Zustand, Redux, or React Query). - Create reusable components for chat UI, system prompt editor, model selection dropdown, settings panel. - Improve TTS with pluggable drivers. - Persist data (settings, saved prompts) in local storage or database. - Add hotkeys/keyboard navigation. - Implement dark/light mode. --- ### ✅ **5. Output Formatting for Your Response** Your response **must**: - Output only the final prompt text to be given to the second LLM. - That final prompt should be clearly written as a command/instruction to generate the **CLine prompt**. - Do **not add explanations, commentary, or markdown outside of the final prompt text**. --- **Now generate the final prompt that will be given to the second LLM.** --- Would you like me to now generate that final prompt for you?