V3 Deep Integration

Migrates claude-flow onto agentic-flow@alpha adapters to eliminate duplicate orchestration code.

1|Updated Jun 3, 2026
One-click install
npx skills add https://github.com/KentwareDemo/RuView --skill v3-deep-integration-kentwaredemo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: V3 Deep Integration
Source: https://github.com/KentwareDemo/RuView/tree/main/.claude/skills/v3-integration-deep
Command: npx skills add https://github.com/KentwareDemo/RuView --skill v3-deep-integration-kentwaredemo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? claude-flow duplicates thousands of lines of orchestration logic (swarm coordination, agent management, task scheduling) that already exist in agentic-flow@alpha, creating maintenance burden and missing performance gains from SONA, Flash Attention, and AgentDB. ## Core Features & Use Cases - Adapter Layer Migration: Extends agentic-flow@alpha Agent classes with backward-compatible wrappers so existing claude-flow APIs keep working during transition. - Feature Integration: Wires in SONA learning modes, Flash Attention (2.49x-7.47x speedup), AgentDB HNSW search (150x-12,500x), and 213 MCP tools with 19 hook types. - Phased Code Removal: Replaces SwarmCoordinator, AgentManager, and TaskScheduler implementations, cutting orchestration code from 15,000+ lines to under 5,000. - Use Case: A maintainer planning the v3 architecture runs the three-phase migration (adapter layer, system migration, cleanup) while validating feature parity at each step. ## Quick Start Ask the agent to design the agentic-flow@alpha adapter layer and plan the phased migration of swarm coordination, agent management, and task execution.

Frequently Asked Questions about V3 Deep Integration

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I migrate claude-flow to agentic-flow@alpha?

Follow the three-phase plan: build an adapter layer extending agentic-flow Agent classes, migrate systems one by one (swarm, agents, tasks), then remove deprecated files. Backward compatibility wrappers keep old APIs working during the transition.

How much duplicate code does the v3 integration remove?

The migration targets reducing orchestration code from over 15,000 lines to under 5,000. Major removals include SwarmCoordinator (800+ lines), AgentManager (1,736+ lines), and TaskScheduler (500+ lines), which overlap 50-80% with agentic-flow equivalents.

What performance gains come from agentic-flow@alpha integration?

Flash Attention targets 2.49x-7.47x speedup with 50-75% memory reduction, AgentDB HNSW indexing targets 150x-12,500x search improvement, and SONA real-time mode adapts in roughly 0.05ms.

Does the migration preserve backward compatibility with v2?

Yes, through a gradual strategy: dual operation of old and new systems, feature-by-feature migration with parity validation, then complete transition only after full parity is confirmed. The goal is 100% v2 functionality maintained.

What MCP tools and hooks become available after integration?

The integration exposes 213 pre-built MCP tools and 19 hook types from agentic-flow@alpha. Claude-flow-specific tools are registered on top, and hooks are configured for claude-flow workflows.