// AI Stack dataset: categories, item schema, and curated initial items // This file is self-contained to avoid modifying global types. export type AiCategory = | 'method' | 'ide' | 'assistant' | 'rule' | 'mcp' | 'extension' | 'infra' | 'evaluation' export type AiGroup = | 'ide_dev' | 'protocols' | 'concepts' | 'approaches' export const AI_GROUPS: readonly AiGroup[] = [ 'ide_dev', 'protocols', 'concepts', 'approaches' ] as const export type AiItemLinkset = { setup?: string rules?: string example?: string } export type AiItem = { id: string name: string category: AiCategory group: AiGroup icon?: string shortWhy?: string links?: AiItemLinkset tags?: string[] featured?: boolean heat?: number } export const AI_CATEGORIES: readonly AiCategory[] = [ 'method', 'ide', 'assistant', 'rule', 'mcp', 'extension', 'infra', 'evaluation' ] as const export const aiStackItems: AiItem[] = [ // Concepts and methodologies { id: 'method-bmad', name: 'BMAD Method', category: 'method', group: 'concepts', icon: 'i-twemoji-jigsaw', shortWhy: 'Outcome-driven orchestration with one-tool-per-step discipline', tags: ['bmad', 'workflow', 'governance'] }, { id: 'method-openspec', name: 'Spec-Driven Development', category: 'method', group: 'concepts', icon: 'i-twemoji-open-book', shortWhy: 'Develop AI systems guided by specification before execution', tags: ['spec', 'planning'] }, { id: 'concepts-hitl', name: 'Human-In-The-Loop', category: 'method', group: 'concepts', icon: 'i-twemoji-handshake', shortWhy: 'Integrate human oversight to improve outputs and control model decisions', tags: ['concept', 'quality'] }, { id: 'concepts-agent-coordination', name: 'Agent-Based Coordination', category: 'method', group: 'concepts', icon: 'i-twemoji-people-holding-hands', shortWhy: 'Coordinate multiple agents to accomplish complex goals', tags: ['agents', 'coordination'] }, { id: 'concepts-token-budget', name: 'Token Budget Management', category: 'method', group: 'concepts', icon: 'i-twemoji-abacus', shortWhy: 'Strategically manage context and token usage for LLMs', tags: ['tokens', 'context'] }, // IDE / Development Environments { id: 'ide-vscode', name: 'VSCode', category: 'ide', group: 'ide_dev', icon: 'i-logos-visual-studio-code', shortWhy: 'Popular, professional IDE for coding and AI projects', tags: ['ide', 'vscode'] }, { id: 'ide-kiro', name: 'Kiro', category: 'ide', group: 'ide_dev', icon: 'i-twemoji-brain', shortWhy: 'Intelligent editor and environment for AI software development', tags: ['ide', 'kiro'] }, { id: 'ide-cursor', name: 'Cursor', category: 'ide', group: 'ide_dev', icon: 'i-twemoji-sparkles', shortWhy: 'AI-native IDE with rules and repo-aware edits', tags: ['ide', 'cursor'] }, { id: 'ide-roocode', name: 'RooCode', category: 'ide', group: 'ide_dev', icon: 'i-twemoji-robot', shortWhy: 'Auto-planning agents for repo-wide diffs', tags: ['ide', 'agents'] }, { id: 'ide-qoder', name: 'Qoder', category: 'ide', group: 'ide_dev', icon: 'i-twemoji-rocket', shortWhy: 'Team collaboration and fast AI-powered coding environment', tags: ['ide', 'qoder'] }, // Protocols & Standards (MCP) { id: 'protocols-mcp', name: 'MCP (Machine Collaboration Protocol)', category: 'mcp', group: 'protocols', icon: 'i-twemoji-globe-with-meridians', shortWhy: 'Collaboration and coordination protocol for multi-agent AI projects', tags: ['mcp', 'protocol'] }, { id: 'mcp-context7', name: 'Context7 MCP', category: 'mcp', group: 'protocols', icon: 'i-twemoji-open-book', shortWhy: 'Retrieve up-to-date docs and examples for libraries', tags: ['mcp', 'context7'] }, { id: 'mcp-playwright', name: 'Playwright MCP', category: 'mcp', group: 'protocols', icon: 'i-logos-playwright', shortWhy: 'Browser automation and testing via MCP-compatible server', tags: ['mcp', 'playwright'] }, { id: 'mcp-chrome-devtools', name: 'Chrome DevTools MCP', category: 'mcp', group: 'protocols', icon: 'i-logos-chrome', shortWhy: 'Control Chrome DevTools for inspection and performance via MCP', tags: ['mcp', 'chrome'] }, { id: 'mcp-browsertools', name: 'BrowserTools MCP', category: 'mcp', group: 'protocols', icon: 'i-twemoji-globe-with-meridians', shortWhy: 'General browser tools MCP for automation and scraping', tags: ['mcp', 'browser'] }, // Emerging approaches { id: 'approaches-context-management', name: 'Context Management', category: 'method', group: 'approaches', icon: 'i-twemoji-card-index-dividers', shortWhy: 'Control and retain context and memory for long-running agent interactions', tags: ['context', 'memory'] }, { id: 'approaches-agent-orchestration', name: 'Agent Orchestration', category: 'method', group: 'approaches', icon: 'i-twemoji-ringed-planet', shortWhy: 'Coordinate groups of agents to solve multi-step or complex tasks', tags: ['agents', 'orchestration'] }, { id: 'approaches-cot', name: 'Chain-of-Thought Reasoning', category: 'method', group: 'approaches', icon: 'i-twemoji-thought-balloon', shortWhy: 'Step-by-step reasoning and analysis with language models', tags: ['reasoning'] }, { id: 'approaches-prompt-engineering', name: 'Prompt Engineering Tools/Techniques', category: 'method', group: 'approaches', icon: 'i-twemoji-hammer-and-wrench', shortWhy: 'Craft and evaluate prompts to achieve precise, targeted model outputs', tags: ['prompting'] } ]