cot-widget-spike
CoT Reasoning Map
从 Claude 的思维链中学习。一个用于 Claude Code 的 MCP 应用 (SEP-1865),它将扩展思维痕迹转化为交互式推理地图——类型化移动(框架、假设、验证、回溯、自我修正、洞察、行动、结论),以带有依赖和丢弃边的图形、努力时间线、回放模式以及点击跳转到逐字 CoT 摘录的形式呈现。
无需 API 密钥:您正在运行的 Claude 会话即可完成注释。
工作原理
list_sessions— 查找包含思考块的近期 Claude Code 转录文件 (~/.claude/projects/**/*.jsonl)get_thinking— 从会话中提取编号的思考块 + 工具调用骨架调用中的 Claude 将思考注释为带有逐字摘录的类型化移动
show_reasoning_map— 将移动渲染为聊天中的内联交互式小部件(Claude Code 桌面应用 / claude.ai;终端 CLI 中的文本回退)
可以说类似:"映射我上次会话的推理" 或 "给我展示一个演示推理地图"(无参数调用会渲染示例数据)。
Related MCP server: Sequential Thinking MVP Server
安装
需要 Node 18+。
git clone https://github.com/matanma12/cot-widget-spike.git
cd cot-widget-spike
npm install && npm run build
claude mcp add cot-spike --scope user -- node "$(pwd)/server.js"重启 Claude Code 并请求一个推理地图。
开发
server.js— MCP 服务器(stdio;npm run serve:http用于在 :3001 上提供 Streamable HTTP)mcp-app.html+src/mcp-app.js— 小部件(来自@modelcontextprotocol/ext-apps的App)npm run build— 打包到dist/mcp-app.html(vite-plugin-singlefile)node test-client.js— 通过 stdio 的协议冒烟测试
状态 / 路线图
已完成:数据摄取、宿主模型注释循环、图形 + 时间线 + 回放 + 摘录、在 Claude Code 桌面中验证的内联渲染。下一步:通过钩子实时跟踪会话、跨痕迹的模式库、预测下一步移动的回放、claude.ai 部署。
MIT 许可。
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseBquality-maintenanceProvides structured sequential thinking capabilities for AI assistants to break down complex problems into manageable steps, revise thoughts, and explore alternative reasoning paths.29
- AlicenseBquality-maintenanceEnables AI assistants to perform structured, step-by-step reasoning by breaking down complex problems into numbered thoughts, with support for revising previous steps and exploring alternative reasoning paths.5
- AlicenseAqualityBmaintenanceEnables structured step-by-step reasoning with branching, revisions, and self-critique to help break down complex problems into manageable steps with confidence tracking and thought history search.7416MIT
- AlicenseBqualityDmaintenanceProvides 10 structured reasoning strategies (Chain of Thought, ReAct, Tree of Thoughts, etc.) for complex problem-solving with session persistence, branching, and tool integration capabilities.32127MIT
Related MCP Connectors
Cross-agent artifact workspace with provenance across Claude Code, Codex, Cursor, LangGraph.
Let Claude, Cursor, or ChatGPT author Mermaid diagrams your team can read and share.
AI Agent with Architectural Memory. Impact analysis (free), tests and code from the graph (pro).
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/matanma12/cot-widget-spike'
If you have feedback or need assistance with the MCP directory API, please join our Discord server