/migrate gaps — Show Discovery Gaps & Assumption Validation

Show unanswered questions, formal assumptions with validation status, and missing discovery dimensions, ranked by their impact on estimate accuracy and confidence score.

Instructions

1. Load Data

1. Call the MCP tool get_assessment with project_path set to the current working directory. If it returns null, fall back to reading .migration/assessment.json. Verify an assessment exists.
2. Call the MCP tool get_discovery with the assessment ID to get all discovery data. If it returns null, fall back to reading ALL files in .migration/discovery/.
3. Read skills/migrate-knowledge/discovery/discovery-tree.json — the full question set
4. Read skills/migrate-knowledge/discovery/dimension-descriptions.md — for context on each dimension
5. Call the MCP tool get_analysis with the assessment ID. If it returns null, fall back to reading .migration/analysis.json. The analysis response includes assumptions data.
6. Call the MCP tool get_estimate with the assessment ID. If it returns null, fall back to reading .migration/estimate.json.

2. Identify Gaps

Categorize gaps into three types:

Missing Dimensions

Dimensions with no discovery file at all (status: not_started). For each, describe:

• What the dimension covers
• Why it matters for estimation
• Default assumptions that will be used if it remains unanswered
• Impact on estimate accuracy (high/medium/low)

Partial Dimensions

Dimensions where status is partial. For each:

• List the specific unanswered required questions
• List any conditional questions that couldn't be evaluated
• Impact on estimate accuracy

Low-Confidence Answers

Answers where confidence is "unknown" or "assumed". For each:

• What was assumed or marked unknown
• What impact the wrong assumption would have
• How to validate (who to ask, what to check)

3. Show Formal Assumptions (if registry exists)

If assumptions-registry.json is available, present assumptions with their formal tracking data:

For each assumption, show:

• ID: ASMP-NNN
• Value: What's currently assumed
• Basis: Why this value was assumed
• Confidence: assumed or unknown
• Validation Status: unvalidated / validated / invalidated
• Affected Components: Which estimate components this impacts
• Pessimistic Widening: How many hours this adds to the pessimistic range
• Validation Method: How to confirm this assumption

Sort assumptions by pessimistic_widening_hours (highest impact first).

4. Impact Assessment

Rate each gap's impact on estimate accuracy:

High Impact — Could change the total estimate by >20%:

• Database HA configuration unknown
• xConnect/xDB scope unclear
• EXM enablement unknown
• Number of custom integrations unknown
• Networking/VPN requirements unknown

Medium Impact — Could change the estimate by 10-20%:

• Exact database sizes unknown
• Solr configuration details missing
• CDN complexity unknown
• Monitoring requirements unspecified
• Backup/DR requirements unclear

Low Impact — Changes estimate by <10%:

• Exact certificate count unknown
• DNS record count unknown
• Specific CI/CD tooling unknown
• Monitoring dashboard details missing

5. Calculate Estimate Tightening Effect

If both assumptions-registry.json and estimate.json exist, calculate the effect of validating assumptions:

For each unvalidated assumption (sorted by impact):

1. Calculate the pessimistic range reduction: pessimistic_widening_hours would be removed
2. Calculate the new confidence score if this assumption were validated
3. Show cumulative effect of validating the top N assumptions

Present as:

ESTIMATE TIGHTENING PROJECTIONS
═══════════════════════════════

Current Confidence: 72% | Pessimistic Range: 612 hrs

Validate these assumptions to tighten the estimate:

 #  | Assumption ID | Description                  | Range Reduction | New Confidence
----|---------------|------------------------------|-----------------|---------------
 1  | ASMP-003      | Database size (<50GB)         | -16 hrs         | 76%
 2  | ASMP-007      | xDB contact count (<100K)     | -12 hrs         | 79%
 3  | ASMP-011      | Custom integration count (3)   | -8 hrs          | 82%
    |               |                              |                 |
    | CUMULATIVE    | Validate all 3 above          | -36 hrs         | 82%

Validating these 3 assumptions reduces pessimistic range by 36 hours
and improves confidence from 72% to 82%.

6. Present Results

Display gaps organized by impact, incorporating formal assumption IDs:

Discovery Gaps & Assumptions Analysis
═════════════════════════════════════

Completeness: 12 of 17 dimensions complete (71%)
Confidence Score: 72% (12 assumptions, 3 validated)

HIGH IMPACT GAPS (affect estimate by >20%)
──────────────────────────────────────────

1. [ASMP-007] xConnect contact volume [assumed: <100K]
   Basis: Inferred from XP Scaled default
   Validation: Query xDB collection database for contact count
   If wrong: +36-48 hrs | Pessimistic widening: +12 hrs
   Affects: xconnect_xdb, database_single

2. Networking/Firewall [not started]
   Related: ASMP-009, ASMP-010, ASMP-011
   Missing: VPC layout, subnet design, VPN/ExpressRoute needs
   Impact: Cannot size networking phase. Could add 20-60 hours.

MEDIUM IMPACT GAPS
──────────────────
...

VALIDATED ASSUMPTIONS (confirmed)
─────────────────────────────────
✓ ASMP-001: CM instance count = 1 (confirmed by client)
✓ ASMP-004: SQL collation = SQL_Latin1_General_CP1_CI_AS
...

7. Assumption Validation Workflow

When the user wants to validate an assumption, follow this workflow:

1. User provides confirmed value — e.g., "ASMP-007 is actually 2.5M contacts"
2. Update via MCP (primary): Call the update_assumption MCP tool with the assessment ID, assumption ID, validation_status ("validated" or "invalidated"), and optionally actual_value. The tool returns the updated confidence score and remaining widening hours.
3. Update JSON snapshot: Also update .migration/assumptions-registry.json:
   • Set validation_status: "validated" (if confirmed as assumed) or validation_status: "invalidated" (if different)
   • If invalidated, record actual_value and update assumed_value
   • Recalculate confidence_score in summary
   • Remove pessimistic_widening_hours for validated assumptions
4. If invalidated, warn about estimate impact:
   • "ASMP-007 invalidated: assumed <100K contacts, actual is 2.5M. This triggers the xdb_very_large multiplier (2.0×) — re-run /migrate estimate to recalculate."
5. Update discovery file — update the corresponding discovery answer with confidence: "confirmed" and the actual value (both via save_discovery MCP tool and .migration/discovery/<dim>.json)

8. Suggest Next Steps

Based on the gaps found:

• If high-impact gaps exist: Suggest specific questions to ask the client, specific things to check in the AWS environment
• If formal assumptions exist: Highlight the top 3 assumptions to validate for maximum confidence improvement
• If only medium/low gaps: Suggest proceeding with /migrate estimate with noted assumptions
• If no gaps: Congratulate the user on thorough discovery and suggest /migrate estimate or /migrate plan
• If estimate exists: Show the tightening projections and suggest validating specific assumptions

Offer to:

• Re-enter discovery mode for specific dimensions: "Want me to run through the networking questions? Run /migrate discover"
• Validate specific assumptions: "Tell me the confirmed value for any assumption ID and I'll update the registry"
• Re-run estimate after validation: "After validating assumptions, run /migrate estimate to see the tightened range"