Recursive Thinking MCP Server
思维链递归思维 MCP 服务器
v2.2.0 - 专家角色、自定义检查清单和严格验证!
高效的 Token MCP 服务器,通过结构化推理链、自我批判和自动重思循环,使 AI 智能体能够进行更深入的思考。
🎯 v2.2.0 新特性
特性 | 描述 | 优势 |
专家角色 | AI 扮演特定角色(例如:安全架构师) | 基于视角的严谨思考 |
自定义检查清单 | 根据特定要求验证解决方案 | 定制的质量执行 |
严格模式 | 除非满足所有标准,否则禁止定稿 | 高标准的生产质量 |
思维链 | AI 解释选择每个解决方案的原因 | 透明的推理,更好的决策 |
自我评估 | AI 对自己的解决方案进行评分(1-10 分) | 对解决方案质量的诚实评估 |
重思循环 | 如果评分 <7 或发现问题,自动进行重思 | 持续改进 |
Related MCP server: Sequential Thinking MCP Server
✨ 特性
🧠 思维链:AI 必须解释每个决策背后的推理
🎭 专家角色:将专业知识注入思维过程
📋 自定义检查清单:强制 AI 根据您的特定要求进行验证
⚖️ 严格模式:确保 AI 绝不满足于“足够好”
📊 自我评估:AI 使用分数(1-10)、好/坏标记和理由来评估解决方案
🔄 持续重思:如果解决方案评分 <7 或存在弱点,自动进行重思
🎯 基于阶段的工作流:结构化的 5 阶段思维过程
💰 Token 优化:智能压缩,专注于阶段的提示词(减少约 70%)
🔄 工作原理
5 个阶段
┌─────────────┐
│ 1. EXPLORE │ ──→ Propose solution with reasoning & expert perspective
└──────┬──────┘
▼
┌─────────────┐
│ 2. JUSTIFY │ ──→ Critically examine reasoning & assumptions
└──────┬──────┘
▼
┌─────────────┐
│ 3. EVALUATE│ ──→ Rate solution (1-10) against checklist
└──────┬──────┘
▼
┌───┴───┐
│ Score │
│ ≥7? │ ──NO─→ ┌─────────────┐
└───┬───┘ │ 4. RETHINK │ ──→ Improve based on weaknesses
│YES └──────┬──────┘
▼ │
┌─────────────┐ │
│ 5. FINALIZE│ ←──────────┘
└─────────────┘🚀 使用方法
使用高级配置启动思维会话
await mcp.callTool('recursive_thinking', {
action: 'start',
task: 'Design a highly available database schema for a global payment system',
config: {
role: 'Database Reliability Engineer',
checklist: [
'Must handle 10k transactions/sec',
'ACID compliance mandatory',
'Multi-region failover strategy',
'Data encryption at rest and in transit'
],
strictMode: true,
minRating: 8
}
});⚙️ 配置选项
参数 | 类型 | 默认值 | 描述 |
| string | - | 要扮演的专家角色(例如:'高级 DevOps') |
| string[] | [] | 要验证的具体要求 |
| boolean | false | 如果为 true,则在未满足标准时失败而不是定稿 |
| number | 5 | 最大递归深度 |
| number | 0.85 | 当置信度达到阈值时停止 |
| number | 7 | 最低可接受分数(1-10) |
| boolean | true | 启用持续优化循环 |
📦 安装
npx -y recursive-thinking-mcp📄 许可证
MIT
🔗 链接
This server cannot be deployed
Maintenance
Related MCP Connectors
Build, validate, and deploy multi-agent AI solutions from any AI environment.
Reduces AI Agent token usage by 40% via three-stage SOP workflow.
Create and manage AI agents that collaborate and solve problems through natural language interacti…
Verified 2-3 step AI-agent missions with whole-transaction success-only charging.
Related MCP Servers
- AlicenseAqualityCmaintenanceEnables multi-level reasoning with configurable depth levels, session-based state management, structured thought input, and real-time trace resources.1148 npm1MIT
- FlicenseAqualityDmaintenanceEnables AI agents to perform dynamic and reflective problem-solving through a chain of thoughts, allowing them to break down complex problems, revise past thoughts, and explore logic branches before reaching a conclusion.1187,522 npm-
- AlicenseNot gradedqualityCmaintenanceProvides persistent cross-session AI memory, code graph analysis, and subagent context compression so coding agents can resume work instantly and avoid re-reading the codebase.Elastic 2.0
- AlicenseNot gradedqualityCmaintenanceEnables agents to reason through a validated 28-layer cognitive substrate, evolve tree-of-thoughts memory across sessions, and communicate with peer agents via A2A tools.16 PyPI1MIT