Evaluation Habits and the Team AI Working Agreement

Path Outline

Progress0%

1The AI Draft–Human Decide Operating Model for Scrum 2The AI Collaboration Loop and Artifact Validation Checks 3Data-Handling Guardrails for Scrum Inputs 4LLM Workflow Threats in Everyday Scrum Work 5Lightweight Prompting Patterns for Reliable Drafts

6Evaluation Habits and the Team AI Working Agreement

7A Safe Refinement Input Bundle for AI-Assisted Backlog Work 8AI-Drafted PBIs With Human-Owned Value and Priority 9A Story Quality Checklist for Ambiguity, Scope, and Dependencies 10Story Splitting Options That Preserve Value and Learning 11Acceptance Criteria Drafting and Validation for Testability 12Acceptance Criteria, Definition of Done, and the Refinement Decision Log

13Sprint Planning Outputs and Decision Rights With AI in the Room 14AI-Drafted Sprint Goal and Scope Options With Explicit Assumptions 15Capacity Reality Checks That Prevent AI Overcommit 16Dependency and Risk Blind Spots in AI-Generated Sprint Plans 17Task-Level Plans as Drafts: Sequencing for Integration and Done 18Uncertainty-First Forecasting and the Overcommit Correction Log

19The Daily Scrum Boundary: Planning Signal vs Status Reporting 20A Privacy-Safe Daily Update Input Format for AI Summaries 21Prompting AI for a Sprint-Goal-Linked Daily Summary Pack 22Human Validation: Accuracy, Attribution, Tone, and Sensitive Content 23Action Logs, Exceptions, and Sharing Norms

24AI’s Role Boundary in the Sprint Retrospective 25Traceable Theme Clustering From Raw Retro Notes 26From Themes to Owned, Testable Experiments 27Sentiment as a Weak Signal With Psychological Safety Constraints 28Forecast Ranges From Sprint History With Explicit Uncertainty 29Regime-Change Overrides and the End-to-End AI Assist Audit Trail

Hide outline

Home
Paths
AI-Assisted Scrum: Draft With AI, Decide as a Team
Evaluation Habits and the Team AI Working Agreement