CLIne prompt for documentation

Using these repos in the root I would like to rebuilt the chrisbot. I want to use the customgpt-starter-kit to help with components, I want to use the next folder's boilerplate as a starting point I want to use TTS for the text to speech for the application. I want to use graphRAG or at least just normal RAG on the information in the data folder. I want to only use one model for the inference which is the llama.cpp mlabbone_gemma-3 gguf in the root. Please help me draft a prompt for CLIne which will generate in the already made docs folder all of the documentation you would need for such a project including architecture, API documentation, mapping functions and dependencies, explaining the application in detail, as well as all of the standards for programming and testing and debugging and how to compose a checklist and ledger to keep track of progress as well as generate a first class README.md for the github repo that is world class.

Here’s a **complete, well-structured prompt you can give to CLIne** to begin generating all the documentation and standards you requested for rebuilding the Chrisbot project using your repos, tech stack, and constraints. --- ### ✅ **CLIne Prompt (Copy & Use)** **System Instructions for CLIne — Generate Full Documentation for Chrisbot Rebuild** You are an elite AI software architect and technical writer. Your task is to generate, inside the existing `/docs` folder at the root of the project, all of the documentation required to build a production-grade version of the Chrisbot application using the following requirements: --- ### **📁 Project Context & Repos** The root directory contains these relevant folders and files: - `/next` → Next.js frontend boilerplate (to be used as the UI foundation) - `/customgpt-starter-kit` → For reusable UI components, chat UI structures, API utilities - `/data` → All knowledge sources for RAG or GraphRAG - `/models/mlabbone_gemma-3.gguf` → The **only model used for inference** via llama.cpp - `/chrisbot` or root files → legacy bot logic to be modernized - `/docs` → Location where all documents must be created and organized --- ### **🎯 Application Goals** Generate documentation to build a chatbot application that: - Uses **Next.js** frontend from the `/next` folder - Imports UI/logic patterns from `customgpt-starter-kit` - Runs inference locally using **llama.cpp with model mlabbone_gemma-3.gguf** - Implements **TTS (Text-to-Speech)** for AI-generated responses (e.g., Edge-TTS or any local TTS) - Supports **RAG or GraphRAG** on `/data` contents - Is modular, production-ready, testable, and well-documented --- ### **📚 Required Documents to Generate in `/docs`** Create the following files: 1. **`/docs/ARCHITECTURE.md`** - System overview & diagrams - Frontend → Backend → Inference pipeline - RAG/GraphRAG data flow - TTS integration pipeline - API call lifecycle (from message → model → response) 2. **`/docs/API_REFERENCE.md`** - All REST/WebSocket endpoints - Input/output schema for chat, RAG, TTS request - Error responses & rate limiting strategy 3. **`/docs/COMPONENT_MAP.md`** - List each module/function/class - Where it lives in the repo - Dependencies and data flow between components - How Next.js pages interact with customgpt-starter-kit and llama.cpp 4. **`/docs/RAG_IMPLEMENTATION.md`** - Basic RAG pipeline - Optionally GraphRAG setup - File ingestion strategy for `/data` - Vector DB or local embeddings storage method 5. **`/docs/PROGRAMMING_STANDARDS.md`** - Coding style (TypeScript, Node, Python bindings if needed) - Folder structure standards - Naming conventions - Git commit message conventions 6. **`/docs/TESTING_AND_DEBUGGING.md`** - Unit test patterns (Jest, Vitest, etc.) - Integration testing approach for AI inference - Logging, tracing, error boundaries - How to debug llama.cpp and TTS issues 7. **`/docs/PROJECT_MANAGEMENT.md`** - How to build a task checklist - Progress ledger format - Example Kanban board layout - Milestones to complete the MVP 8. **✅ Generate a world-class `README.md` at root** - Project overview - Tech stack - Setup and run instructions - Screenshots / demo structure - Contribution guidelines - Badges and license section placeholders --- ### **⚙ Technical Constraints** - **Only use `mlabbone_gemma-3.gguf` via llama.cpp** for model inference - Use **TTS** (specify command, library, or API used) - Use **Next.js App Router** - Local-only, no external OpenAI/Anthropic APIs - All documentation must be production-worthy and well-structured in Markdown --- ### **✅ Output Format** - Generate **all files directly into `/docs` and `README.md`** - Use clean, well-formatted Markdown - Include tables, diagrams (Mermaid), and code samples where helpful - Do not write code implementation yet — only documentation & structure --- **Now begin generating the documentation.** --- ### ✅ Would You Like Me To Automatically Trigger This in CLIne Format? I can format it as a JSON command for CLIne if needed. Want that next?