Pattern: Advisory Debate | Team size: 5 agents
This team evaluates competing priorities through structured debate, balancing HR operations, new hire experience, manager needs, compliance, and technical feasibility. The advisory debate pattern works well because onboarding dashboards often fail when one perspective dominates and practical tradeoffs are not tested early.
I need to build a HR onboarding dashboard for new hires.
Create an agent team to design a practical HR onboarding dashboard for new hires using the “Employee Experience War Room” advisory debate pattern. Build a complete, self-contained set of planning deliverables that balances new hire experience, HR operations, hiring manager needs, compliance/privacy, and technical feasibility. Project name: hr_onboarding_dashboard Create all outputs under this exact directory: outputs/agent_teams_demo/hr_onboarding_dashboard/ Business problem: The organization needs to build an HR onboarding dashboard for new hires. The dashboard should help new hires understand what to do before day one, during the first week, and through the first 30/60/90 days. It should also help HR teams administer onboarding, help hiring managers track readiness and assign tasks, and ensure privacy, compliance, accessibility, and auditability. Assume the dashboard may serve: - New hires, including employees who have accepted offers but have not started yet - HR onboarding coordinators - Hiring managers - IT, facilities, payroll, benefits, and learning/training stakeholders as task owners - Compliance/audit reviewers with limited access Assume likely capabilities include: - Personalized onboarding checklist - Required forms and document collection - E-signature/document acknowledgement status - Training assignments - First-week schedule - Benefits and payroll setup reminders - Equipment/access readiness - Manager task assignments - Escalations for overdue tasks - HR admin view - Notifications and reminders - Role-based access control - Basic reporting and audit trail Do not design UI mockups. Focus on dashboard requirements, scope, workflows, tradeoffs, risks, and implementation-ready product direction. Use the following agent team: 1. Employee Advocate Role: Assess every dashboard feature from the new hire perspective to ensure clarity, confidence, accessibility, and low friction. Primary lens: New hire anxiety, clarity, inclusive language, mobile access, cognitive load, first-day confidence, accessibility, self-service. Must challenge: HR-heavy workflows, confusing status labels, excessive notifications, irrelevant information, hidden dependencies. 2. HR Operations Advisor Role: Ensure the dashboard supports real HR onboarding processes, ownership rules, escalations, and administrative workflows. Primary lens: Process ownership, operational scalability, task templates, exceptions, handoffs, HRIS integration, lifecycle stages. Must challenge: Features that look good but require manual HR work, unclear accountability, duplicate systems of record, non-scalable exceptions. 3. Hiring Manager Advisor Role: Represent manager needs for visibility, reminders, task assignments, approvals, and team readiness. Primary lens: Manager accountability, readiness tracking, team introductions, equipment/access status, role-specific preparation, lightweight approvals. Must challenge: Dashboards that exclude managers, overload managers with admin work, fail to show blockers, or create unclear ownership. 4. Compliance and Privacy Reviewer Role: Challenge proposed features against data privacy, document retention, labor policy, accessibility, and audit requirements. Primary lens: Data minimization, consent, retention, RBAC, audit trail, legal hold, accessibility, cross-border privacy considerations. Must challenge: Unnecessary sensitive data exposure, weak permissions, unlogged changes, ambiguous document storage, inaccessible UX assumptions. 5. Product Decision Facilitator Role: Synthesize debate outcomes into prioritized dashboard requirements, tradeoff decisions, and a practical MVP scope. Primary lens: Prioritization, MVP definition, sequencing, decision clarity, stakeholder alignment, implementation feasibility. Must challenge: Overly broad scope, unresolved conflicts, vague requirements, missing acceptance criteria, features with unclear users. Collaboration mechanics: - Use a structured advisory debate. Each agent must first produce an independent memo from their assigned perspective. - After all independent memos are complete, each agent must read the other agents’ memos and identify at least 5 points of agreement, 5 tensions, and 3 assumptions that need validation. - Agents must directly challenge assumptions, not personalities. Use evidence, operational logic, user risk, compliance risk, or feasibility impact. - The Compliance and Privacy Reviewer has veto authority over features that create unacceptable legal, privacy, retention, accessibility, or audit risk unless mitigations are documented. - The Product Decision Facilitator must not erase disagreement. Capture unresolved tradeoffs explicitly and recommend a practical decision. - Every major recommendation must state which user group benefits, what problem it solves, and what risk or cost it introduces. - Always end with a synthesis/review step that validates the full output for completeness, contradictions, missing dependencies, and MVP clarity. Execution sequence and dependencies: Step 1 — Create project brief. This task must complete before any agent memo begins. Create outputs/agent_teams_demo/hr_onboarding_dashboard/00_project_brief.md Length: 700-1,000 words. Required sections: 1. Problem Statement 2. Target Users 3. Onboarding Timeline Covered 4. Assumed Systems and Integrations 5. Candidate Dashboard Capabilities 6. Key Constraints 7. Open Questions 8. Definition of Success Step 2 — Independent agent memos. These tasks may run in parallel only after 00_project_brief.md is complete. Each memo must be 900-1,200 words. Each memo must include exactly these sections: 1. Perspective Summary 2. Top 10 Dashboard Needs From This Perspective 3. Features to Avoid or Defer 4. Workflow Implications 5. Data, Notification, and Permission Concerns 6. Success Metrics 7. Top Risks 8. Assumptions to Validate 9. Non-Negotiable Requirements Create these files: - outputs/agent_teams_demo/hr_onboarding_dashboard/01_employee_advocate_memo.md - outputs/agent_teams_demo/hr_onboarding_dashboard/02_hr_operations_advisor_memo.md - outputs/agent_teams_demo/hr_onboarding_dashboard/03_hiring_manager_advisor_memo.md - outputs/agent_teams_demo/hr_onboarding_dashboard/04_compliance_privacy_reviewer_memo.md Agent-specific requirements: Employee Advocate memo: - Include at least 5 recommendations that reduce new hire anxiety. - Include at least 4 accessibility or inclusion considerations. - Include at least 3 examples of dashboard language that should be simplified. - Include recommendations for pre-day-one, day-one, first-week, and first-30-days views. HR Operations Advisor memo: - Include an ownership model covering HR, manager, IT, facilities, payroll, benefits, learning/training, and employee. - Include at least 6 administrative workflow requirements. - Include escalation rules for overdue tasks. - Include template and exception-handling requirements. Hiring Manager Advisor memo: - Include at least 8 manager-specific dashboard requirements. - Include guidance for task assignment, reminders, approvals, and team readiness. - Include a manager view for pre-start, first week, and first 30/60/90 days. - Include risks of manager overload and how to mitigate them. Compliance and Privacy Reviewer memo: - Include at least 10 compliance/privacy/accessibility requirements. - Include role-based access rules for new hire, HR, manager, IT/facilities, payroll/benefits, learning/training, and auditor. - Include document retention and audit trail expectations. - Include a “blocked unless mitigated” list of risky features. Step 3 — Cross-agent debate. This task must complete after all four independent memos are complete and before synthesis begins. Create outputs/agent_teams_demo/hr_onboarding_dashboard/05_advisory_debate_log.md Length: 1,200-1,600 words. Required sections: 1. Debate Method 2. Shared Agreements 3. Major Tensions 4. Feature-by-Feature Debate Table 5. Assumptions Challenged 6. Compliance Vetoes or Conditional Approvals 7. Decisions Made 8. Issues Requiring Product Leadership Decision The Feature-by-Feature Debate Table must include at least 15 rows and these columns: - Candidate Feature - Employee Advocate View - HR Operations View - Hiring Manager View - Compliance/Privacy View - Decision - Rationale The Major Tensions section must include at least these debates: - Personalization vs. privacy/data minimization - Comprehensive dashboard vs. cognitive overload - Manager visibility vs. employee confidentiality - Automated reminders vs. notification fatigue - HR admin flexibility vs. process standardization - Pre-start access vs. identity/security constraints Step 4 — Prioritized requirements. This task must complete after 05_advisory_debate_log.md. Create outputs/agent_teams_demo/hr_onboarding_dashboard/06_prioritized_requirements.md Length: 1,500-2,000 words. Required sections: 1. Product Goal 2. Personas and Permission Levels 3. MVP Requirements 4. Post-MVP Requirements 5. Explicitly Deferred Requirements 6. Functional Requirements 7. Non-Functional Requirements 8. Data and Integration Requirements 9. Notification Requirements 10. Reporting and Audit Requirements 11. Acceptance Criteria 12. Dependencies and Constraints Requirements format: - Include at least 25 total requirements. - Label each requirement as P0, P1, or P2. - For each requirement include: - Requirement ID - Requirement statement - Primary user - Priority - Rationale - Acceptance criteria - Key dependency - Risk if omitted MVP minimum scope must include: - New hire personalized checklist - Timeline segmented by pre-start, day one, first week, first 30 days - Required documents/forms status without exposing sensitive content unnecessarily - Training/task assignments - First-week schedule or orientation events - Manager readiness view - HR admin task/status view - Notifications/reminders - Role-based access control - Audit trail for status changes and document acknowledgements - Accessibility requirements Step 5 — Workflow and operating model. This task must complete after 06_prioritized_requirements.md begins but before final synthesis is complete. Create outputs/agent_teams_demo/hr_onboarding_dashboard/07_workflows_operating_model.md Length: 1,200-1,600 words. Required sections: 1. End-to-End Onboarding Workflow 2. New Hire Journey 3. HR Operations Workflow 4. Hiring Manager Workflow 5. Cross-Functional Task Owner Workflow 6. Escalation Model 7. Role and Ownership Matrix 8. Status Model 9. Exception Handling 10. Operational Metrics Include: - A RACI-style ownership matrix with at least 10 onboarding activities. - A status model with at least 8 statuses, including blocked, overdue, completed, waived, and not applicable. - Escalation rules for overdue, blocked, missing owner, failed integration sync, and compliance-critical tasks. - Explicit handoffs between HR, manager, IT/facilities, payroll/benefits, learning/training, and employee. Step 6 — Compliance, privacy, and risk register. This task must complete after 05_advisory_debate_log.md and before final synthesis. Create outputs/agent_teams_demo/hr_onboarding_dashboard/08_compliance_privacy_risk_register.md Length: 1,000-1,400 words. Required sections: 1. Compliance Position 2. Data Classification 3. Role-Based Access Control Rules 4. Document Retention and Deletion Rules 5. Audit Trail Requirements 6. Accessibility Requirements 7. Privacy-by-Design Requirements 8. Risk Register 9. Mitigation Plan 10. Release Blockers Risk Register format: Include at least 12 risks. Each risk must include: - Risk ID - Description - Severity: Low/Medium/High/Critical - Likelihood: Low/Medium/High - Owner - Mitigation - Release impact Include at least 5 release blockers that must be resolved before launch. Step 7 — MVP scope and roadmap. This task must complete after 06_prioritized_requirements.md, 07_workflows_operating_model.md, and 08_compliance_privacy_risk_register.md are complete. Create outputs/agent_teams_demo/hr_onboarding_dashboard/09_mvp_scope_roadmap.md Length: 1,000-1,400 words. Required sections: 1. MVP Scope Summary 2. Included in MVP 3. Excluded From MVP 4. Release 1 Acceptance Criteria 5. 30/60/90-Day Roadmap 6. Implementation Dependencies 7. Validation Plan 8. Success Metrics 9. Go/No-Go Criteria Include: - At least 12 MVP features. - At least 8 deferred features with rationale. - At least 10 measurable success metrics. - A phased roadmap for launch, 30 days post-launch, 60 days post-launch, and 90 days post-launch. - A go/no-go checklist with at least 12 criteria. Step 8 — Final synthesis and review. This task must complete last, after all previous files are complete. Create outputs/agent_teams_demo/hr_onboarding_dashboard/10_final_synthesis_review.md Length: 1,200-1,700 words. Required sections: 1. Executive Summary 2. Final Recommended Dashboard Direction 3. Key Decisions From the Advisory Debate 4. MVP Recommendation 5. Major Tradeoffs and Rationale 6. Open Questions for Stakeholders 7. Contradictions or Gaps Found During Review 8. Final Checklist 9. Next Steps The Final Checklist must verify: - Every agent perspective was represented. - Compliance/privacy risks were addressed. - MVP scope is practical and not overloaded. - Requirements include acceptance criteria. - Workflows have clear owners. - Role-based access is explicit. - Notifications are scoped to avoid fatigue. - Accessibility is included. - Audit trail and retention needs are included. - Dependencies are explicit. - Final outputs do not contradict each other. - Remaining open questions are clearly listed. Quality requirements for all files: - Use clear markdown headings. - Be specific, not generic. - Do not include placeholder text. - Do not invent company-specific legal conclusions; frame uncertain legal areas as items requiring legal review. - Use concise business language suitable for product, HR, engineering, security, and compliance stakeholders. - Make assumptions explicit. - Distinguish MVP from post-MVP. - Include practical implementation implications. - Avoid recommending features that expose sensitive data without a role-based justification. - Ensure every recommendation has a clear user or operational benefit. Final instruction: After creating all files, review the complete set as the Product Decision Facilitator. Confirm that the advisory debate produced a coherent MVP and that all dependencies, risks, tradeoffs, and unresolved stakeholder questions are visible in 10_final_synthesis_review.md.
Create your own AI agent team at Build Agents Store. Describe your business problem and get specialized agent teams with ready-to-use prompts for Claude Code.