Consulting QA Skill

{WORKSPACE} = timestamped run directory provided by the orchestrator (e.g., _workspace/run_20260404_153000).

Purpose

Every persuasion scenario must pass quality validation through 5 distinct consulting methodologies before reaching the final report. This skill provides the evaluation framework, grading criteria, and revision standards.

When to Use

• Persuasion scenarios need quality validation
• Arguments need to be stress-tested from multiple perspectives
• Defense strategies need professional-grade quality assurance
• Need to identify which arguments are too weak for client-facing use

Two-Phase QA Architecture

QA operates in TWO sequential phases. Phase A is the most critical.

Phase A: Foundational Audit (Premise Validation)

Runs FIRST, before any argument grading. Validates that the entire upstream pipeline produced correct foundations.

If ANY foundational check fails → set the appropriate REDO verdict, write detailed foundation_issues, and DO NOT proceed to Phase B. The orchestrator will restart the pipeline from the failing phase.

Phase B: Argument Quality (Scenario Quality Validation)

Only runs if Phase A verdict is FOUNDATION_PASS.

The Five Lenses — Quick Reference

Read references/big5-consulting-methods.md for detailed evaluation criteria, real-world examples, and grading rubrics for each firm.

Grading Scale

Verdict Logic

IF any_lens_grade == "F" → FAIL
IF count(grade <= "D") >= 2 → FAIL
IF any_lens_grade == "D" → REVISION_NEEDED
IF cultural_calibration_failed → REVISION_NEEDED
IF consistency_check_failed → REVISION_NEEDED
ELSE → PASS

Cross-Scenario Consistency Rules

Arguments across different findings must not contradict each other. Common contradictions to catch:

1. Speed vs. Thoroughness: One argument claims "we prioritized speed" while another claims "we were thorough"
2. Cost vs. Quality: One claims "cost-optimized" while another claims "premium quality"
3. Innovation vs. Precedent: One says "novel approach" while another says "proven methodology"
4. Scope: Arguments about what's in/out of scope must be consistent

Cultural Calibration Checklist

For each scenario, verify:

• No direct confrontation or accusation
• Acknowledgment precedes every counter-argument
• Humble language used consistently
• Face-saving bridge is genuinely face-saving (not condescending)
• Japanese phrasing is grammatically correct and appropriately formal (敬語/丁寧語)
• Nemawashi recommendations are realistic for the context
• No Western-style "hard sell" or aggressive persuasion

Revision Instructions Format

When writing revision instructions, be specific:

Bad: "Strengthen the evidence in Level 2" Good: "Level 2 cites 'industry data' without specifics. Replace with: cite the 2024 JFSA (金融庁) guidelines Section 4.2 which mandates minimum 99.95% uptime for Tier-1 banking systems, directly supporting our 99.99% SLA commitment."

Process

1. Read all upstream artifacts (01, 02, 03 JSON files)
2. For each scenario, apply all 5 lenses
3. Grade each level within each lens
4. Check cross-scenario consistency
5. Run cultural calibration checklist
6. Assign verdict (PASS/REVISION_NEEDED/FAIL)
7. Write specific revision instructions for non-PASS scenarios

Output Format

Write QA results to {WORKSPACE}/04_qa_results.json. See the agent definition for the complete schema.

Quality Checklist

• All 5 lenses applied to every scenario
• Level-by-level grades provided (not just overall)
• Specific improvement actions for every grade below B
• Cross-scenario consistency verified
• Cultural calibration checked
• Revision instructions are actionable and specific
• Overall defense readiness score calculated