.kiro/skills/qe-aqe-v2-v3-migration/SKILL.md
Migrate Agentic QE projects from v2 to v3 with zero data loss
npx skillsauth add proffesor-for-testing/agentic-qe qe-aqe-v2-v3-migrationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
<default_to_action> When migrating from v2 to v3:
Never delete v2 data without explicit user confirmation. </default_to_action>
# When v3 becomes main release, just update the package
npm install agentic-qe@latest
# Run migration
aqe migrate
# Or use this skill
/aqe-v2-v3-migration
| Component | v2 Location | v3 Location | Auto-Migrate |
|-----------|-------------|-------------|--------------|
| Memory DB | .agentic-qe/memory.db | .aqe/agentdb/ | Yes |
| Config | .agentic-qe/config.json | .aqe/config.json | Yes |
| Patterns | .agentic-qe/patterns/ | .aqe/reasoning-bank/ | Yes |
| Cache | .agentic-qe/cache/ | .aqe/cache/ | Optional |
| Logs | .agentic-qe/logs/ | .aqe/logs/ | No (fresh start) |
.agentic-qe/ directory)aqe --version (should be 2.x.x)npm run backup (in v2 project)npm install agentic-qe@latestaqe migrateaqe testaqe coverageaqe patterns listv2 Structure: v3 Structure:
├── src/mcp/tools/ ├── src/domains/
│ ├── test-*.ts (40+ tools) │ ├── test-generation/
│ └── ... │ ├── test-execution/
├── src/core/agents/ │ ├── coverage-analysis/
│ ├── mixed agents │ ├── quality-assessment/
│ └── ... │ ├── defect-intelligence/
└── src/core/memory/ │ ├── requirements-validation/
└── scattered impls │ ├── code-intelligence/
│ ├── security-compliance/
│ ├── contract-testing/
│ ├── visual-accessibility/
│ ├── chaos-resilience/
│ └── learning-optimization/
├── src/kernel/
│ ├── event-bus.ts
│ └── coordinator.ts
└── src/mcp/
└── domain-handlers.ts
| v2 API | v3 API | Notes |
|--------|--------|-------|
| aqe init | aqe init | Different binary |
| aqe.generateTests() | testGeneration.generate() | Domain-based |
| aqe.analyzeGaps() | coverageAnalysis.findGaps() | O(log n) now |
| memory.store() | agentDB.store() | HNSW-indexed |
| patterns.learn() | reasoningBank.record() | With verdicts |
{
"version": "2.8.2",
"memory": {
"path": ".agentic-qe/memory.db",
"type": "sqlite"
},
"agents": {
"enabled": ["test-generator", "coverage-analyzer"]
}
}
{
"version": "3.0.0",
"kernel": {
"eventBus": "in-memory",
"coordinator": "queen"
},
"domains": {
"test-generation": { "enabled": true },
"test-execution": { "enabled": true },
"coverage-analysis": {
"enabled": true,
"algorithm": "hnsw",
"dimensions": 128
}
},
"memory": {
"backend": "agentdb",
"path": ".aqe/agentdb/",
"hnsw": {
"M": 16,
"efConstruction": 200
}
},
"learning": {
"reasoningBank": true,
"sona": true
}
}
// v2: Direct SQLite access
import Database from 'better-sqlite3';
const db = new Database('.agentic-qe/memory.db');
const patterns = db.prepare('SELECT * FROM patterns').all();
// v3: AgentDB with HNSW
import { AgentDB } from 'agentic-qe';
const db = new AgentDB('.aqe/agentdb/');
await db.initialize({ dimensions: 128, M: 16 });
// Migration script transfers and indexes
for (const pattern of v2Patterns) {
await db.store({
key: pattern.id,
value: pattern.data,
embedding: await generateEmbedding(pattern.data),
metadata: { migratedFrom: 'v2', originalId: pattern.id }
});
}
Import Paths
// v2
import { AgenticQE } from 'agentic-qe';
// v3 (when v3 becomes main release, package name is still 'agentic-qe')
import { TestGenerationDomain } from 'agentic-qe/domains';
CLI Commands
# v2
aqe test --parallel
# v3
aqe test --workers=4 --topology=mesh
MCP Server
# v2
claude mcp add aqe -- npx aqe-mcp
# v3 (same CLI name, enhanced capabilities)
claude mcp add aqe -- npx aqe mcp
aqe.runTests() → Use domain-specific methodsIf migration fails or you need to revert:
# 1. v3 does NOT modify v2 data
# Your .agentic-qe/ folder is untouched
# 2. Downgrade to v2
npm install [email protected]
rm -rf .aqe/
# 3. Continue using v2
aqe --version # Should show 2.x.x
// Use Task tool to spawn migration agents in parallel
Task({
prompt: "Analyze v2 memory.db and extract all patterns",
subagent_type: "researcher",
description: "Analyze v2 patterns"
});
Task({
prompt: "Convert v2 config to v3 format",
subagent_type: "coder",
description: "Convert config"
});
Task({
prompt: "Validate migration results",
subagent_type: "tester",
description: "Validate migration"
});
| Issue | Cause | Solution |
|-------|-------|----------|
| "Cannot find .agentic-qe/" | No v2 installation | Run aqe init first |
| "Memory migration failed" | Corrupted SQLite | Use backup: npm run backup:restore |
| "HNSW index error" | Dimension mismatch | Set dimensions: 128 in config |
| "Pattern not found" | Not migrated | Re-run: aqe migrate --patterns |
# Run migration with debug output
DEBUG=aqe:migrate aqe migrate
# Check migration logs
cat .aqe/logs/migration.log
[v2-v3-migration] tag| v2 Version | v3 Version | Migration Support | |------------|------------|-------------------| | 2.8.x | 3.0.x | Full | | 2.7.x | 3.0.x | Full | | 2.6.x | 3.0.x | Partial (config only) | | 2.5.x and below | 3.0.x | Manual migration |
Skill Version: 1.0.0 | Last Updated: 2026-01-11
development
Apply XP practices including pair programming, ensemble programming, continuous integration, and sustainable pace. Use when implementing agile development practices, improving team collaboration, or adopting technical excellence practices.
development
Warehouse Management System testing patterns for inventory operations, pick/pack/ship workflows, wave management, EDI X12/EDIFACT compliance, RF/barcode scanning, and WMS-ERP integration. Use when testing WMS platforms (Blue Yonder, Manhattan, SAP EWM).
testing
Advanced visual regression testing with pixel-perfect comparison, AI-powered diff analysis, responsive design validation, and cross-browser visual consistency. Use when detecting UI regressions, validating designs, or ensuring visual consistency.
development
Comprehensive truth scoring, code quality verification, and automatic rollback system with 0.95 accuracy threshold for ensuring high-quality agent outputs and codebase reliability.