feat: add multi-agent team memory support (v2.0)

- Add agent ID schema with actor attribution and scope filtering
- Implement ScopedMemoryManager for dual-layer memory (team + agent)
- Add team configuration file _omp/config.yaml
- Implement handoff workflow for agent-to-agent task transfer
- Update templates with multi-agent support documentation
- Extend xmemory-types with MemoryScope, AgentDefinition, HandoffRecord
- Update README with v2.0 features

New types:
- MemoryScope: 'team' | 'agent'
- AgentDefinition: id, role, channels, description
- AgentRole: coordinator, architect, developer, etc.
- TeamConfig: team name, runId, agents array
- ActorAttribution: actorId, actorType, timestamp
- HandoffRecord: fromAgent, toAgent, context, progress

Architecture:
- Team shared memory: _omp/memory/
- Agent private memory: _omp/agents/{agent_id}/memory/
- Handoff records: _omp/memory/handoffs/

Author: Alenryuichi

README.md

@@ -868,17 +868,23 @@ npx openmemory-plus install
 ### 💡 Later (未来考虑)
 
 - [ ] Web UI 管理界面
-- [ ] 团队记忆共享 (可选)
 - [ ] 云端同步选项
 - [ ] 记忆分析和洞察
 
 👉 **[查看完整 Roadmap Issues](https://github.com/Alenryuichi/openmemory-plus/issues?q=is%3Aissue+is%3Aopen+label%3A%22roadmap%3A+now%22%2C%22roadmap%3A+next%22%2C%22roadmap%3A+later%22)**
 
-### ✅ Done (v1.6)
+### ✅ Done (v2.0)
 
 <details>
 <summary>已完成功能</summary>
 
+**v2.0 - Multi-Agent Team Support**
+- [x] 🤝 多 Agent 团队记忆架构 (Team Shared + Agent Private)
+- [x] 🎭 Agent ID Schema (actor attribution, scope filtering)
+- [x] 📋 团队配置文件 (`_omp/config.yaml`)
+- [x] 🔄 Handoff 工作流 (Agent 间任务交接协议)
+- [x] 📁 Scoped Memory API (双层作用域记忆)
+
 **v1.6 - xMemory 四层架构**
 - [x] 🧠 xMemory 四层架构 (L3 Theme → L2 Semantic → L1 Episode → L0 Message)
 - [x] 🎯 L3 主题层自动聚类 (吸附/分裂/合并算法)

README_EN.md

@@ -695,7 +695,21 @@ The wizard will:
 
 ## 🗺️ Roadmap
 
-### v1.5 (Current)
+### v2.0 (Current) - Multi-Agent Team Support
+- [x] 🤝 Multi-Agent team memory architecture (Team Shared + Agent Private)
+- [x] 🎭 Agent ID Schema (actor attribution, scope filtering)
+- [x] 📋 Team configuration file (`_omp/config.yaml`)
+- [x] 🔄 Handoff workflow (agent-to-agent task transfer protocol)
+- [x] 📁 Scoped Memory API (dual-layer scoped memory)
+
+### v1.6 - xMemory 4-Layer Architecture
+- [x] 🧠 xMemory 4-layer architecture (L3 Theme → L2 Semantic → L1 Episode → L0 Message)
+- [x] 🎯 L3 theme layer auto-clustering (attach/split/merge algorithms)
+- [x] 🔍 Top-down adaptive retrieval (`--level theme`)
+- [x] ⚡ Theme search CLI options (`--no-expand`)
+- [x] 📊 Connected component clustering (xMemory paper implementation)
+
+### v1.5 - Foundation
 - [x] Dual-layer memory architecture
 - [x] Smart classification routing
 - [x] Multi-IDE support (Augment, Claude, Cursor, Gemini)
@@ -706,18 +720,11 @@ The wizard will:
 - [x] Progressive config (append to existing files)
 - [x] MCP auto-configuration and verification
 
-### v2.0 (Planned)
-- [ ] Web UI management interface
-- [ ] Team memory sharing
-- [ ] Memory import/export
-- [ ] Custom classification rules
-- [ ] More IDE support (Windsurf, Cline)
-
 ### v3.0 (Future)
+- [ ] Web UI management interface
 - [ ] Cloud sync option
 - [ ] Memory analytics and insights
 - [ ] AI-driven memory organization
-- [ ] Enterprise features
 
 ---
 

_bmad-output/planning-artifacts/research/technical-multi-agent-team-memory-system-research-2026-03-01.md

@@ -0,0 +1,41 @@
+---
+stepsCompleted: []
+inputDocuments: []
+workflowType: 'research'
+lastStep: 1
+research_type: 'technical'
+research_topic: 'multi-agent-team-memory-system'
+research_goals: '为多 Agent 团队（如 openclaw）找到最佳记忆系统方案，支持团队共享记忆、私有记忆、上下文传递'
+user_name: 'Ryuichialen'
+date: '2026-03-01'
+web_research_enabled: true
+source_verification: true
+---
+
+# Research Report: Technical Research
+
+**Date:** 2026-03-01
+**Author:** Ryuichialen
+**Research Type:** Technical - Multi-Agent Team Memory System
+
+---
+
+## Research Overview
+
+本研究旨在为多 Agent 团队协作场景找到最佳的记忆系统方案。
+
+**研究背景：**
+- 用户有一个 openclaw agent 团队（main, tech-lead, dev-bot, algo-bot, devops-bot）
+- 需要支持团队共享记忆 + 私有记忆 + Agent 间上下文传递
+- 现有 OpenMemory Plus (xMemory 4层架构) 需要增强以支持多 Agent 场景
+
+**研究方法：**
+- 深度网络搜索现有方案
+- 多源验证关键技术主张
+- 架构模式分析
+- 与 OpenMemory Plus 整合可行性评估
+
+---
+
+<!-- Content will be appended sequentially through research workflow steps -->
+

cli/package.json

@@ -1,6 +1,6 @@
 {
   "name": "openmemory-plus",
-  "version": "1.5.0",
+  "version": "2.0.0",
   "description": "Agent Memory Management CLI - Install, configure, and manage OpenMemory Plus",
   "type": "module",
   "bin": {

cli/src/lib/memory/index.ts

@@ -2,6 +2,7 @@
  * OpenMemory Plus - Memory Module
  * Phase 2: Enhanced Memory Capabilities
  * Phase 3: xMemory Hierarchical Integration
+ * Phase 4: Multi-Agent Team Support
  */
 
 export * from './types.js';
@@ -14,3 +15,6 @@ export * from './filesystem.js';
 export * from './theme-manager.js';
 export * from './hierarchy-graph.js';
 export * from './adaptive-search.js';
+
+// Multi-Agent Team Support (Phase 4)
+export * from './scoped-memory.js';