mirror of
https://github.com/proffesor-for-testing/agentic-qe.git
synced 2026-09-19 08:45:47 +08:00
Profa
49e4b4b95c
V3 Skills Improvement Initiative: Agent Naming Standardization (ADR-037): - Updated v3-qe-fleet-coordination to use v3-qe-* prefix - Updated v3-qe-mcp to use v3-qe-* prefix - Updated v3-qe-integration to use v3-qe-* prefix - Migrated 32 agent references from V2 to V3 naming New Enhanced Skills: - v3-qe-memory-unification: AgentDB + HNSW (150x-12,500x faster) - v3-qe-mcp-optimization: Connection pooling, O(1) lookup, <100ms targets - v3-qe-agentic-flow-integration: SONA, Flash Attention (2.49x-7.47x), 9 RL algorithms New ADRs (ADR-037 to ADR-041): - ADR-037: V3 QE Agent Naming Standardization - ADR-038: V3 QE Memory System Unification - ADR-039: V3 QE MCP Optimization - ADR-040: V3 QE Agentic-Flow Integration - ADR-041: V3 QE CLI Enhancement Documentation: - Created docs/v3/skills-improvement-plan.md - Updated v3-adrs.md with all new ADRs (now 41 total) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
5.7 KiB
5.7 KiB
V3 QE Skills Improvement Plan
Date: 2026-01-11 Status: In Progress Author: Claude Code
Executive Summary
This document outlines the plan to improve V3 QE skills by:
- Updating agent references from V2 to V3 naming conventions
- Creating enhanced QE-specific versions of claude-flow V3 skills
- Aligning with existing V3 QE agent definitions
Current State Analysis
Existing V3 QE Skills (21 skills)
- v3-qe-memory-system
- v3-qe-mcp
- v3-qe-integration
- v3-qe-core-implementation
- v3-qe-cli
- v3-qe-fleet-coordination
- v3-qe-learning-optimization
- v3-qe-ddd-architecture
- v3-qe-test-generation
- v3-qe-test-execution
- v3-qe-coverage-analysis
- v3-qe-quality-assessment
- v3-qe-defect-intelligence
- v3-qe-code-intelligence
- v3-qe-security-compliance
- v3-qe-security
- v3-qe-contract-testing
- v3-qe-chaos-resilience
- v3-qe-requirements-validation
- v3-qe-visual-accessibility
- v3-qe-performance
Existing V3 QE Agents (47 agents)
Located in .claude/agents/v3/, agents use v3-qe-* prefix:
- v3-qe-test-architect
- v3-qe-tdd-specialist
- v3-qe-coverage-specialist
- v3-qe-quality-gate
- v3-qe-defect-predictor
- v3-qe-parallel-executor
- v3-qe-learning-coordinator
- v3-qe-flaky-hunter
- v3-qe-security-scanner
- ... and 38 more
Claude-Flow V3 Skills (12 skills)
- v3-core-implementation
- v3-ddd-architecture
- v3-memory-unification
- v3-mcp-optimization
- v3-integration-deep
- v3-cli-modernization
- v3-performance-optimization
- v3-swarm-coordination
- v3-security-overhaul
Issues Identified
1. Agent Naming Inconsistency
Problem: V3 QE skills reference V2-style short agent names instead of V3 full names.
Affected Skills:
| Skill | V2 References | V3 Names |
|---|---|---|
| v3-qe-fleet-coordination | 'test-architect' |
'v3-qe-test-architect' |
| v3-qe-fleet-coordination | 'coverage-specialist' |
'v3-qe-coverage-specialist' |
| v3-qe-fleet-coordination | 'quality-gate' |
'v3-qe-quality-gate' |
| v3-qe-fleet-coordination | 'defect-predictor' |
'v3-qe-defect-predictor' |
| v3-qe-mcp | 'test-architect' |
'v3-qe-test-architect' |
| v3-qe-mcp | 'tdd-specialist' |
'v3-qe-tdd-specialist' |
| v3-qe-integration | 'test-architect' |
'v3-qe-test-architect' |
2. Missing Enhanced Skills
Gap Analysis: Claude-flow V3 skills have features not present in QE equivalents:
| Claude-Flow Skill | QE Equivalent | Missing Features |
|---|---|---|
| v3-memory-unification | v3-qe-memory-system | ADR-006/009 alignment, data migration |
| v3-mcp-optimization | v3-qe-mcp | Connection pooling, load balancing, <100ms targets |
| v3-integration-deep | v3-qe-integration | Deep agentic-flow integration, SONA/Flash Attention |
| v3-cli-modernization | v3-qe-cli | Interactive prompts, workflow orchestration |
Improvement Plan
Phase 1: Agent Naming Updates (Priority: High)
ADR-037: V3 QE Agent Naming Standardization
Update all V3 QE skills to use consistent V3 agent naming:
// Before
agents: ['test-architect', 'coverage-specialist']
// After
agents: ['v3-qe-test-architect', 'v3-qe-coverage-specialist']
Files to Update:
.claude/skills/v3-qe-fleet-coordination/SKILL.md.claude/skills/v3-qe-mcp/SKILL.md.claude/skills/v3-qe-integration/SKILL.md.claude/skills/v3-qe-learning-optimization/SKILL.md
Phase 2: Enhanced Memory System (Priority: High)
ADR-038: V3 QE Memory System Unification
Create v3-qe-memory-unification skill with:
- AgentDB with HNSW indexing (150x-12,500x faster search)
- Migration from legacy QE memory systems
- Cross-domain memory sharing for 12 DDD domains
- SONA integration for pattern learning
- Memory performance benchmarks
Phase 3: MCP Optimization (Priority: Medium)
ADR-039: V3 QE MCP Optimization
Create v3-qe-mcp-optimization skill with:
- Connection pooling for QE MCP tools
- O(1) tool lookup via hash table
- Load balancing for fleet operations
- <100ms response time targets
- Performance monitoring dashboard
Phase 4: Agentic-Flow Integration (Priority: Medium)
ADR-040: V3 QE Agentic-Flow Deep Integration
Create v3-qe-agentic-flow-integration skill with:
- SONA learning mode integration for QE
- Flash Attention (2.49x-7.47x speedup) for QE workloads
- Code deduplication strategy
- 8 RL algorithm integration
Phase 5: CLI Enhancement (Priority: Low)
ADR-041: V3 QE CLI Enhancement
Enhance v3-qe-cli with:
- Interactive test generation wizard
- Progress bars for fleet operations
- Workflow automation for QE pipelines
- Intelligent command completion
Implementation Timeline
| Phase | ADR | Start | Duration |
|---|---|---|---|
| 1 | ADR-037 | Immediate | 1 day |
| 2 | ADR-038 | After Phase 1 | 2 days |
| 3 | ADR-039 | After Phase 1 | 2 days |
| 4 | ADR-040 | After Phase 2 | 3 days |
| 5 | ADR-041 | After Phase 3 | 2 days |
Success Metrics
- All V3 QE skills use V3 agent naming
- v3-qe-memory-unification skill created and validated
- v3-qe-mcp-optimization skill created with <100ms targets
- v3-qe-agentic-flow-integration skill created
- v3-qe-cli enhanced with interactive features
- All ADRs approved and documented
- Integration tests passing with real agents