📄️ Overview
Structured Responsible AI assessment through a 6-phase conversational workflow aligned with Microsoft RAI Standard v2 and NIST AI RMF 1.0
📄️ Why RAI Planning?
How structured Responsible AI assessment catches risks that traditional security planning misses and why it matters for AI systems
📄️ Agent Architecture
Technical architecture of the RAI Planner agent including state management, instruction file references, interaction model, and session recovery
📄️ Entry Modes
Three ways to start a RAI assessment with the RAI Planner agent including capture, from-prd, and from-security-plan modes
📄️ Phase Reference
Complete specification of the RAI Planner's six assessment phases including NIST AI RMF mapping, artifacts, and state transitions
📄️ Handoff Pipeline
How the RAI Planner generates scorecard results, produces backlog items, and connects to the Security Planner through the from-security-plan entry mode