ruflo Agent 能力体系能力矩阵、claude-flow CLI 查询与 Swarm 角色注册表实战指南【免费下载链接】ruflo The original agent meta-harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, RAG integration, and native Claude Code / Codex / Hermes and many more Integrated项目地址: https://gitcode.com/GitHub_Trending/cl/ruflo本文围绕 rufloclaude-flow的 Agent 能力模型展开先完整解读官方能力矩阵Capability Matrix中四类核心 Agent 的技能划分与适用场景再结合 CLI 源码中的能力注册表、agent_spawnMCP 工具链与 Agent YAML 配置说明能力标签在 Agent 创建、元数据传递和团队编成中的实际流转过程。读完后你将掌握如何用npx claude-flow agents capabilities查询能力、如何为 Swarm 选配合适的 Agent 类型以及能力标签背后的源码级实现依据。一、能力矩阵四类核心 Agent 的技能划分ruflo 的 Agent 能力体系以能力矩阵为骨架。官方文档 .claude/commands/agents/agent-capabilities.md 定义了如下矩阵描述了每类 Agent 的主技能Primary Skills与最佳适用场景Best ForAgent TypePrimary SkillsBest ForcoderImplementation, debuggingFeature developmentresearcherAnalysis, synthesisRequirements gatheringtesterTesting, validationQuality assurancearchitectDesign, planningSystem architecture这张矩阵传递的核心思想是每个 Agent 类型对应一组离散的能力标签标签决定了它在 Swarm 中的分工边界——coder实现与调试承接功能开发覆盖从需求到可运行代码的实现链路researcher分析与综合负责需求收集、资料调研与信息综合tester测试与验证承担质量保证负责测试用例执行与结果验证architect设计与规划负责系统架构层面的设计与总体规划。在 ruflo 的Agent meta-harness定位中见 README这种按能力划分的 Agent 类型是编排多智能体 Swarm、协调自主工作流的基础单元——编排器依据类型与能力标签把任务路由给合适的 Agent而不是让每个 Agent 承担所有职责。二、查询能力claude-flow CLI 命令文档给出了两条核心查询命令# List all capabilities npx claude-flow agents capabilities # For specific agent npx claude-flow agents capabilities --type coder第一条不带参数列出所有 Agent 类型及其能力第二条通过--type coder过滤到指定类型只展示该类 Agent 的能力标签。需要说明的是适用前提文档中的命令走npx claude-flow这一 npm 包入口。从当前仓库源码结构看CLI 命令树中可直接确认的 Agent 命令为claude-flow agent单数形式其子命令包括spawn / list / status / stop / metrics / pool / health / logs / wasm-*等见 agent.ts 中agentCommand的subcommands与帮助输出约 L1080-L1113。agents capabilities查询与agent list、spawn 结果输出共同构成查看能力信息的入口实际可用子命令请以你所安装版本的--help输出为准。三、源码纵深能力注册表 getAgentCapabilities文档矩阵是精选视图而 CLI 内部维护了一份更完整的能力注册表。在 agent.ts 末尾的getAgentCapabilities()辅助函数L1117-L1131中可以看到九种 Agent 类型到能力标签数组的完整映射function getAgentCapabilities(type: string): string[] { const capabilities: Recordstring, string[] { coder: [code-generation, refactoring, debugging, testing], researcher: [web-search, data-analysis, summarization, citation], tester: [unit-testing, integration-testing, coverage-analysis, automation], reviewer: [code-review, security-audit, quality-check, documentation], architect: [system-design, pattern-analysis, scalability, documentation], coordinator: [task-orchestration, agent-management, workflow-control], security-architect: [threat-modeling, security-patterns, compliance, audit], memory-specialist: [vector-search, agentdb, caching, optimization], performance-engineer: [benchmarking, profiling, optimization, monitoring] }; return capabilities[type] || [general]; }对比文档矩阵与源码注册表可以读出三层信息能力标签是细粒度的技能原语文档矩阵中的 Implementation, debuggingcoder在源码中被展开为code-generation / refactoring / debugging / testing四个标签Testing, validationtester展开为unit-testing / integration-testing / coverage-analysis / automation。矩阵是给人看的抽象注册表是给编排器消费的精确标签。注册表比矩阵更广除矩阵中的四类源码还内置了reviewer、coordinator、security-architect、memory-specialist、performance-engineer等类型覆盖代码评审、多智能体编排、安全架构、AgentDB 记忆检索与性能工程等场景。未知类型有安全兜底return capabilities[type] || [general]保证任何未注册类型至少获得[general]能力不会在 spawn 流程中抛错——从源码结构看这是一个面向扩展的宽容设计。四、AGENT_TYPES可 spawn 的完整 Agent 类型清单与能力注册表配套的是 agent.ts 中的AGENT_TYPES常量L52 起它定义了交互式选择与--type参数校验的全部合法取值const AGENT_TYPES [ { value: coder, label: Coder, hint: Code development with neural patterns }, { value: researcher, label: Researcher, hint: Research with web access and data analysis }, { value: tester, label: Tester, hint: Comprehensive testing with automation }, { value: reviewer, label: Reviewer, hint: Code review with security and quality checks }, { value: architect, label: Architect, hint: System design with enterprise patterns }, { value: coordinator, label: Coordinator, hint: Multi-agent orchestration and workflow }, { value: analyst, label: Analyst, hint: Performance analysis and optimization }, { value: optimizer, label: Optimizer, hint: Performance optimization and bottleneck analysis }, { value: security-architect, label: Security Architect, hint: Security architecture and threat modeling }, { value: security-auditor, label: Security Auditor, hint: CVE remediation and security testing }, { value: memory-specialist, label: Memory Specialist, hint: AgentDB unification (150x-12,500x faster) }, { value: swarm-specialist, label: Swarm Specialist, hint: Unified coordination engine }, { value: performance-engineer, label: Performance Engineer, hint: 2.49x-7.47x optimization targets }, { value: core-architect, label: Core Architect, hint: Domain-driven design restructure }, { value: test-architect, label: Test Architect, hint: TDD London School methodology } ];spawn子命令中--type / -t选项直接以AGENT_TYPES.map(a a.value)作为合法取值集合choices——也就是说CLI 会在参数层面阻止输入未定义的类型若不带--type且处于交互模式则弹出上述 15 个选项供选择见 spawn 命令 action 中的select分支。这解释了文档矩阵为何只列 4 类矩阵面向如何分工的选型教学而 choices 校验面向实际能创建什么的完整能力面。五、能力标签的运行时流转spawn → MCP agent_spawn → 元数据能力标签不是静态文档而是在 Agent 生命周期中真实流转的运行时数据。以claude-flow agent spawn -t coder --name bot-1为例agent.ts 的 spawn actionL147-L185中发生了三步流转第一步调用 MCP 工具创建 Agent并把能力写入元数据。const result await callMCPTool{ agentId: string; agentType: string; ... }( agent_spawn, { agentType, id: agentName, config: { provider, model, task, timeout, autoTools }, priority: normal, metadata: { name: agentName, capabilities: getAgentCapabilities(agentType), // 能力注册表在此注入 }, } );第二步MCP 服务端接收capabilities参数。在 AgentTools.ts 中agent_spawn工具的参数 schemaL29-L39明确声明{ name: agent_spawn, description: Spawn a new agent in the swarm, parameters: { type: object, properties: { id: { type: string, description: Unique agent identifier }, type: { type: string, description: Agent type (coder, tester, reviewer, etc.) }, capabilities: { type: array, items: { type: string }, description: Agent capabilities } }, required: [id, type] } }即capabilities是一个可选的字符串数组与id、type一起构成 Swarm 内新成员的身份描述。同一工具族还提供agent_list列出 Swarm 全部 Agent与agent_terminate按agentId终止构成创建—查询—回收的完整闭环。第三步CLI 以表格形式回执能力信息。spawn 成功后输出属性表其中一行专门展示能力{ property: Capabilities, value: getAgentCapabilities(agentType).join(, ) }所以当你执行npx claude-flow agents capabilities --type coder或 spawn 一个 coder 后看到的code-generation, refactoring, debugging, testing都源自同一份注册表——查询命令与运行时元数据保证了一致性。六、Agent 配置文件capabilities 作为持久化字段能力标签还以 YAML 配置的形式持久化在 v3/agents/ 目录下每个 Agent 类型一个文件coder.yaml、architect.yaml、reviewer.yaml、security-architect.yaml、tester.yaml。以 coder.yaml 为例# coder agent configuration type: coder version: 3.0.0 capabilities: - code-generation - refactoring - debugging optimizations: - flash-attention - token-reduction createdAt: 2026-01-13T00:03:03.440Zarchitect.yaml 则为type: architect version: 3.0.0 capabilities: - system-design - api-design - documentation optimizations: - context-caching - memory-persistence从配置文件结构看每个 Agent 定义包含四要素身份typeversion、能力capabilities标签数组、运行时优化策略optimizations如 coder 的flash-attention、token-reductionarchitect 的context-caching、memory-persistence、审计字段createdAt。这意味着能力不仅是 spawn 时的元数据也是可版本化、可审计的静态资产——注意 coder 的 YAML 能力code-generation / refactoring / debugging与 CLI 注册表相比少了testing说明配置文件与运行时注册表是两层独立维护的声明实际能力以具体加载路径为准。七、基于能力矩阵的编队建议结合文档矩阵的 Best For 列与源码中的扩展类型一个典型的 ruflo Swarm 编队可按如下方式组合需求阶段researcherAnalysis, synthesis → Requirements gathering先行收集与分析需求其web-search / contenteditable="false">【免费下载链接】ruflo The original agent meta-harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, RAG integration, and native Claude Code / Codex / Hermes and many more Integrated项目地址: https://gitcode.com/GitHub_Trending/cl/ruflo创作声明：本文部分内容由AI辅助生成（AIGC），仅供参考