Blackboard Architecture
Domain: Agent Architectures
🎯 Motivation
A shared knowledge base (blackboard) that multiple agents read from and write to. Agents contribute partial solutions that build up over time.
📋 Concrete Example
Blackboard: {topic: ‘AI ethics’, research: ’…’, outline: ’…’, draft: ’…’}. Research Agent adds research. Outline Agent reads research, adds outline. Draft Agent reads both, adds draft. All share the same space.
🔗 Analogy
A group whiteboard in a brainstorming session - each person adds ideas, reads others’ ideas, and builds on them. Nobody talks directly; they all write on the board. Blackboard architecture does this for agents.