UNPKG

oneie

Version:

Build apps, websites, and AI agents in English. Zero-interaction setup for AI agents (Claude Code, Cursor, Windsurf). Download to your computer, run in the cloud, deploy to the edge. Open source and free forever.

530 lines (404 loc) 16.3 kB
--- title: One dimension: things category: ONE.md tags: agent, architecture, connections, events, frontend, knowledge, ontology, people, things related_dimensions: connections, events, groups, knowledge, people scope: global created: 2025-11-03 updated: 2025-11-03 version: 1.0.0 ai_context: | This document is part of the things dimension in the ONE.md category. Location: one/things/ONE.md Purpose: Documents one Related dimensions: connections, events, groups, knowledge, people For AI agents: Read this to understand ONE. --- # ONE ██████╗ ███╗ ██╗███████╗ ██╔═══██╗████╗ ██║██╔════╝ ██║ ██║██╔██╗ ██║█████╗ ██║ ██║██║╚██╗██║██╔══╝ ╚██████╔╝██║ ╚████║███████╗ ╚═════╝ ╚═╝ ╚═══╝╚══════╝ Bring your ideas to life https://one.ie npx oneie ## Architecture Overview ONE separates concerns while enabling exponential growth: ``` /Users/toc/Server/ONE/ ├── .claude/ # Shared Claude Code commands, hooks, and agents ├── commands/ # Slash commands: /push-ontology, /push-frontend, etc. └── hooks/ # Shared git hooks and automations ├── one/ # Universal Schema - Ontology documentation (git: one-ie/one) ├── people/ # Individuals, teams, organizations ├── things/ # Products, tools, assets, agents, plans ├── connections/ # Relationships, workflows, integrations ├── events/ # Milestones, triggers, achievements └── knowledge/ # Documentation, architecture, guides ├── web/ # Edge Distribution (git: one-ie/web) ├── backend/ # API Layer (git: one-ie/backend) ├── apps/ # Application repos ├── bullfm/ # (git: bull-fm/bullfm) ├── stack/ # (git: one-ie/stack) ├── one/ # (git: one-ie/one) └── ... └── ONE.md # This file - Architecture documentation ``` ### Installation ```bash # Install CLI globally npm install -g oneie # Or use directly with npx npx oneie init # Creates .claude/ and one/ structure npx oneie clone frontend # Clones one-ie/frontend into /web npx oneie clone backend # Clones one-ie/backend into /backend (later) ``` ### Repository Structure - **Root workspace**: NOT a git repo - just a workspace directory - **Independent repos**: Each subdirectory (`one/`, `web/`, `backend/`) is its own git repo - **Shared tooling**: `.claude/` contains commands and hooks used across all repos - **No submodules**: Simple, elegant - just clone repos where needed ## Repository Map ### Core Systems | Path | Repository | Visibility | Purpose | URLs | | ---------- | --------------------- | ---------- | ---------------------------------------------------- | ------------------ | | `.claude/` | Shared tooling | Private | Claude Code commands, hooks, agents (not a git repo) | - | | `one/` | `one-ie/one-ontology` | Private | Canonical ontology and knowledge graph | - | | `web/` | `one-ie/one` | Public | Demo of ONE experience | https://web.one.ie | | `backend/` | `one-ie/backend` | Private | Knowledge API (Convex + Hono) | https://api.one.ie | ### Applications | Path | Repository | Visibility | Purpose | URLs | | -------------------- | --------------------- | ---------- | ---------------------------------- | --------------------------- | | `apps/bullfm/` | `bull-fm/bullfm` | Private | Bull FM product (fork of frontend) | https://bull.fm | | `apps/stack/` | `one-ie/stack` | Private | Full-stack starter template | https://stack.one.ie | | `apps/astro-shadcn/` | `one-ie/astro-shadcn` | Public | Astro + shadcn showcase | https://astro-shadcn.one.ie | | `apps/astro-email/` | `one-ie/astro-email` | Public | Email toolkit | https://astro-email.one.ie | ### Access Model - **Private**: Ontology (`one/`), backend (`backend/`), Bull FM, Stack starter - **Public**: Web (`web/`), Astro Shadcn, Astro Email - **Shared Tooling**: `.claude/` directory contains commands and hooks used across all repos ### Git Workflow Each repo has its own git remote and can be pushed independently: ```bash # Push ontology changes cd one && git push # Push frontend changes cd web && git push # Push backend changes cd backend && git push # Or use Claude Code commands /push-ontology # Pushes one/ to one-ie/one-ontology /push-web # Pushes frontend/ to one-ie/frontend /push-backend # Pushes backend/ to one-ie/backend ``` ### Why Not Submodules? - **Simplicity**: No submodule complexity, just regular git repos - **Independence**: Each repo can evolve independently - **Flexibility**: Easy to add/remove repos without affecting others - **Tooling**: Claude Code commands provide unified workflow - **Clarity**: Clear separation of concerns, no nested git issues --- ## 1. /one - Universal Schema **The core ontology that connects everything** ### Structure ``` one/ ├── groups/ # Friends or governments ├── people/ # Individuals ├── things/ # Products, tools, assets, resources, technologies ├── connections/ # Relationships, links, associations between entities ├── events/ # Milestones, triggers, moments, achievements └── knowledge/ # Documentation, guides, insights, wisdom ``` ### Philosophy - **Everything is a node**: People, things, events, knowledge - **Everything can connect**: Connections create meaning and value - **Everything evolves**: Events track change and progress over time - **Everything compounds**: Knowledge builds on knowledge exponentially ### Data Model Every entity in `/one` follows this structure: ```yaml entity: id: unique_identifier type: person | thing | connection | event | knowledge metadata: created: timestamp updated: timestamp tags: [] relationships: [] content: markdown_body schema_version: "1.0.0" ``` --- ## 2. /frontend - Edge Distribution **Free ontology framework running at the edge for maximum accessibility** ### Purpose - **Public Knowledge Sharing**: Give away the ontology and methodology for free - **Edge Computing**: Fast, globally distributed via Cloudflare/Vercel edge - **Interactive Experience**: Visualize CASCADE, explore connections, learn methodology - **Open Source**: The framework is the gift; your data stays private ### Tech Stack - **Astro 5**: Static site generation with islands architecture - **Shadcn UI**: Beautiful, accessible component library - **Edge Runtime**: Deploy to Cloudflare Workers or Vercel Edge - **Bun**: Package manager and runtime for development workflows - **Client Integration**: Optionally connects to backend API for authenticated users ### What's Free 1. Vision CASCADE methodology and ontology 2. Test-driven development framework 3. Interactive CASCADE visualizer 4. Public documentation and guides 5. Template structures and schemas 6. Open source agent definitions ### What's Private - Your personal knowledge base (served via backend API) - Your agents and prompts (112+ from ONE-Import) - Your proprietary data and insights - Your company foundation and strategies --- ## 3. /backend - API Layer **Private knowledge API powered by Hono + Convex** ### Purpose - **Data Storage**: Convex database for all entities from `/one` - **API Layer**: Hono HTTP framework for custom routes and logic - **Authentication**: Secure access to private knowledge - **Real-time Sync**: Convex real-time subscriptions - **Edge Compatible**: Can run on Cloudflare Workers ### Tech Stack - **Convex**: Real-time database with built-in auth and queries - **Hono**: Ultra-fast, lightweight HTTP framework for edge - **Bun**: Unified runtime for scripts and tooling - **TypeScript**: Full type safety across frontend/backend - **Edge Deployment**: Runs on Cloudflare Workers alongside frontend ### Schema (Convex) ```typescript // backend/convex/schema.ts // CORE ENTITIES - Map to /one structure people: { id: string name: string role: string bio: markdown connections: string[] // IDs of related entities metadata: object } things: { id: string name: string type: string // tool, product, asset, technology description: markdown connections: string[] metadata: object } connections: { id: string from_entity: string // ID of source entity to_entity: string // ID of target entity relationship_type: string // "uses", "knows", "created", "influences" strength: number // 1-10 metadata: object } events: { id: string type: string // milestone, trigger, achievement entity_id: string // Related entity timestamp: date description: markdown impact: object metadata: object } knowledge: { id: string title: string category: string // guide, insight, playbook, story, mission content: markdown tags: string[] related_entities: string[] metadata: object } // IMPORTED KNOWLEDGE - From ONE-Import agents: { id: string name: string purpose: string capabilities: string[] category: string // marketing, engineering, content, etc. content: markdown quality_gate: string } prompts: { id: string name: string category: string // business, content, marketing content: markdown tags: string[] use_cases: string[] } ontology: { version: string cascade_levels: object // idea→vision→missions→stories→events→tasks test_framework: object metrics: object } ``` ### API Routes (Hono) ```typescript // GET /api/people/:id - Get person by ID // GET /api/things - List all things // GET /api/connections?from=:id - Get entity connections // GET /api/events?entity=:id - Get entity timeline // GET /api/knowledge/search?q=:query - Search knowledge base // GET /api/agents - List all agents // GET /api/agents/:category - Filter by category // GET /api/prompts/:id - Get prompt by ID // GET /api/ontology - Get CASCADE ontology structure ``` --- ## Data Flow ### 1. Knowledge Import (ONE-Import → ONE) ``` ONE-Import ├── /one /Users/toc/Server/ONE/one (Universal Schema) ├── /.claude/agents backend/convex/agents table ├── /one/prompts backend/convex/prompts table ├── /one/ontology.md backend/convex/ontology table └── /apps Reference knowledge, examples ``` ### 2. Frontend Access (Public Edge) ``` User Request Frontend (Edge) ├── Static Pages: Ontology, docs, guides (cached at edge) ├── Interactive Tools: CASCADE visualizer (client-side) └── Dynamic Data: Fetch from backend API (authenticated) ``` ### 3. Backend API (Private Data) ``` API Request Hono Router Convex Database ├── Authentication: Verify user token ├── Query: Fetch data from Convex ├── Transform: Format response └── Return: JSON to frontend or external client ``` --- ## Vision CASCADE Integration The ONE system implements the **Test-Driven Vision CASCADE** methodology: ### 6 Levels of Transformation ```yaml 1. IDEA: Raw possibility Feasibility tests 2. VISION: Future state Alignment validation 3. MISSIONS: Strategic campaigns Success criteria 4. STORIES: Engineering narratives Acceptance testing 5. EVENTS: Milestones & triggers Completion validation 6. TASKS: Concrete actions Quality gates 7. RESULTS: Evaluations 8. PROPERTY: Intellectual property and content, software and data ``` ### How It Maps to ONE - **IDEAS**: Captured in `one/knowledge` as possibilities - **VISIONS**: Defined in `one/knowledge` with alignment tests - **MISSIONS**: Tracked in `one/events` as campaigns - **STORIES**: Documented in `one/knowledge` as narratives - **EVENTS**: Recorded in `one/events` as milestones - **TASKS**: Linked in `one/connections` to `one/people` (agents) ### Agent ONE Orchestration - **Master Orchestrator**: Coordinates 112+ specialized agents - **Test-First**: Every level validated before progression - **Exponential Growth**: 100x multiplication per CASCADE level - **Quality Gates**: 4.0+ stars required at each level --- ## Import Strategy ### Phase 1: Core Structure ✅ - [x] Map to trinity directories (one, web, backend) ### Phase 2: Knowledge Import (Next) ```bash # Import 112+ agents from ONE-Import one-import agents /Users/toc/Server/ONE-Import/.claude/agents # Import 30+ prompts one-import prompts /Users/toc/Server/ONE-Import/one/prompts # Import ontology framework one-import ontology /Users/toc/Server/ONE-Import/one/ontology.md # Import playbooks and guides one-import knowledge /Users/toc/Server/ONE-Import/one/playbooks one-import knowledge /Users/toc/Server/ONE-Import/one/guides # Import company/personal foundation one-import foundation /Users/toc/Server/ONE-Import/one/company one-import foundation /Users/toc/Server/ONE-Import/one/me ``` ### Phase 3: Backend Development ```bash # Extend Convex schema for imported data # Build Hono API routes # Set up authentication and authorization # Deploy to Cloudflare Workers ``` ### Phase 4: Frontend Enhancement ```bash # Build CASCADE visualizer # Create knowledge search interface # Add agent directory browser # Deploy to Cloudflare Pages ``` --- ## Deployment ### Frontend (Edge) ```bash cd web bun install bun run build wrangler pages deploy dist ``` ### Backend (Convex + Hono) ```bash cd backend bunx convex deploy # Hono routes auto-deploy with Convex HTTP functions ``` --- ## Philosophy ### Give Away the Framework, Keep the Knowledge Private - **Open Source**: The ontology, methodology, and CASCADE system - **Free Edge Distribution**: Fast, global access to the framework - **Private Backend**: Your knowledge, agents, and data stay yours - **Value Exchange**: Free framework builds trust and community - **Monetization**: Premium access to your knowledge API, consulting, custom agents ### Edge-First Architecture - **Speed**: Sub-50ms response times globally - **Scale**: Unlimited concurrent users at edge - **Cost**: Near-zero hosting costs with Cloudflare - **Resilience**: Distributed across 200+ data centers - **Privacy**: Backend API separate from public framework ### Test-Driven Everything - **Ideas validated** before becoming visions - **Visions tested** against alignment criteria - **Missions verified** against success metrics - **Stories validated** with acceptance tests - **Events checked** for completion criteria - **Tasks gated** by deliverable quality --- ## Next Steps 1. **Complete Knowledge Import**: Migrate 826 files from ONE-Import 2. **Enhance Backend Schema**: Add tables for all entity types 3. **Build API Routes**: CRUD operations for all entities 4. **Create Frontend UI**: Search, browse, visualize CASCADE 5. **Deploy to Edge**: Cloudflare Pages + Workers 6. **Add Authentication**: Secure private knowledge access 7. **Build Agent System**: 112+ agents accessible via API 8. **Enable Real-time Sync**: Convex subscriptions for live updates --- ## Source Data **Import Source**: `/Users/toc/Server/ONE-Import` a `/Users/toc/Server/ONE/apps/oneie` and bull-fm - 15GB total (826 knowledge files) - 112+ specialized AI agents - 30+ business/content prompts - 20+ example applications - 30 years of accumulated wisdom **Current Home**: `/Users/toc/Server/ONE` - Trinity architecture implemented - Universal schema in `/one` - Edge frontend ready - Backend foundation established --- **ONE System**: Where your 30 years of knowledge becomes exponentially accessible, validated through test-driven CASCADE, and distributed at the edge for maximum impact. _Test-first. Validate continuously. Grow exponentially._