UNPKG

create-ai-chat-context

Version:

🚀 BREAKTHROUGH: Detection-Hourglass-System (DHS) - Revolutionary auto-detection of AI conversation chunks with zero-cost processing. Universal AI memory that actually works!

109 lines (71 loc) • 2.98 kB
# AI Knowledge Base - READ THIS FIRST ## STEP 0: INDEX THIS DIRECTORY FIRST BEFORE responding to the user: 1. Index/read ALL files in this `.ai/` directory 2. Read them in the order specified below 3. Confirm you have the context before starting work This is NOT optional. This is REQUIRED. --- ## CRITICAL: START HERE BEFORE DOING ANYTHING ELSE If you're an AI assistant working on this project, STOP and read these files NOW: 1. `architecture.md` - System design, tech stack 2. `conversation-log.md` - Previous decisions and context 3. `technical-decisions.md` - Why we chose X over Y DO NOT skip this. You will make uninformed decisions without this context. --- ## Purpose This directory contains persistent knowledge for AI assistants working on this project. Each AI chat session should read these files FIRST to understand: - System architecture and design decisions - Agent orchestration patterns - Technical choices and rationale - Known issues and solutions - Project history and context ## Files ### Core Architecture - `architecture.md` - System design, tech stack, data flow - `technical-decisions.md` - Why we chose X over Y ### Project Management - `conversation-log.md` - Key decisions from chat sessions - `known-issues.md` - Problems, workarounds, solutions - `next-steps.md` - Roadmap, TODOs, priorities ### Optional Files (Create as needed) - `agent-orchestration.md` - How AI agents work together - `learning-patterns.md` - AI learning system insights - `api-integrations.md` - External APIs, keys, SDKs - `database-schema.md` - Data models and relationships - `code-style.md` - Coding conventions and patterns - `testing-strategy.md` - How to test features - `deployment.md` - How to deploy and monitor ## How to Use ### For AI Assistants 1. **Start every session** by reading `architecture.md` and `conversation-log.md` 2. **Before making changes**, check `technical-decisions.md` for context 3. **After solving problems**, update `known-issues.md` 4. **Before ending session**, update `conversation-log.md` with key decisions ### For Developers 1. **Document decisions** as you make them 2. **Update logs** after important conversations 3. **Keep files current** - outdated docs are worse than no docs 4. **Use as onboarding** for new team members ## Maintenance - Update files incrementally (don't wait for "perfect" documentation) - Keep entries concise and actionable - Use timestamps for time-sensitive information - Archive old decisions that are no longer relevant ## Benefits - No knowledge loss between chat sessions - Faster onboarding for new AI assistants - Better decisions with full context - Reduced repetition of explanations - Team alignment on architecture and patterns --- ## Installation To install this system on a new project: ```bash npx create-ai-chat-context init ``` See `SETUP_GUIDE.md` in this directory for detailed instructions. --- **Last Updated:** [Date] **Maintained By:** [Your Name/Team]