Designs incremental migration strategies, identifies service boundaries, produces dependency maps and migration roadmaps, and generates API facade designs for aging codebases. Use when modernizing legacy systems, implementing strangler fig pattern or branch by abstraction, decomposing monoliths, upgrading frameworks or languages, or reducing technical debt without disrupting business operations.
94
100%
Does it follow best practices?
Impact
88%
1.18xAverage score across 6 eval scenarios
Passed
No known issues
Strangler fig facade with canary rollout
Facade/router layer
100%
100%
Feature flag routing config
100%
100%
User-hash canary routing
0%
100%
Phased migration plan
100%
100%
Error rate thresholds per phase
100%
100%
Rollback mechanism
100%
100%
Dual-write new-DB-first
0%
100%
Legacy sync non-blocking
0%
100%
Lazy read migration
37%
100%
Legacy cleanup timing
25%
87%
No direct big-bang switch
100%
100%
Characterization tests and branch by abstraction
Characterization tests first
80%
70%
Tests capture current behavior
62%
87%
Abstract interface created
0%
0%
Legacy adapter wraps old code
0%
12%
New implementation alongside legacy
87%
87%
Dependency injection switch
70%
60%
Parallel run comparison
100%
90%
Legacy code retained
100%
100%
Characterization tests still pass
100%
100%
Coverage target noted
0%
100%
No call-site modification before abstraction
16%
25%
Legacy system assessment and migration roadmap
Lines of code analysis
100%
100%
Dependency analysis
83%
100%
Code smell detection
100%
100%
Test coverage estimate
100%
66%
Hotspot identification
50%
66%
SQALE-style tech debt
50%
80%
Risk matrix with scores
100%
100%
Risks prioritized by score
100%
100%
Mitigation per risk
100%
100%
Phased roadmap
100%
100%
Success metrics per phase
100%
100%
Rollback plan per phase
12%
75%
Incremental not big-bang
100%
100%
API versioning migration with deprecation
V1 endpoint preserved
50%
100%
V2 endpoint added
100%
100%
Content negotiation or versioned routing
100%
100%
X-API-Deprecation header
41%
58%
X-API-Sunset header
41%
66%
V1 response structure unchanged
62%
100%
V2 response structure differs
100%
100%
Feature flag or gradual routing
0%
100%
No big-bang removal
100%
100%
Migration documentation
100%
100%
Code output present
100%
100%
Database schema evolution expand-contract
EXPAND step: nullable new column
100%
100%
WRITE BOTH step: dual writes
100%
100%
MIGRATE step: backfill script
100%
100%
READ NEW step: prefer new column
100%
100%
CONTRACT step: drop old column
100%
100%
Steps are separate and ordered
83%
100%
No downtime requirement met
100%
100%
Old column not dropped early
100%
100%
Application model reflects dual state
100%
100%
Migration steps labeled
70%
100%
Migration monitoring, validation, and decision documentation
Error rate metric tracked
100%
100%
Latency metric tracked
100%
100%
Per-phase thresholds defined
100%
100%
Automated rollback trigger
100%
100%
Rollback reduces percentage not kills service
100%
100%
Phase advancement validation
100%
100%
Monitoring dashboard or report
100%
100%
One release cycle at 100% before cleanup
0%
20%
Business behavior validation
0%
25%
Migration decisions documented
100%
100%
Alert notification defined
70%
60%
5b76101
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.