MCP.so
登录

Sequential Thinking MCP Server已精选

@modelcontextprotocol

基本信息

分类

其他

许可证

NOASSERTION

传输方式

stdio

作者

modelcontextprotocol

提交者

idoubi

关于 Sequential Thinking MCP Server

Model Context Protocol 服务器

#sequentialthinking#problem-solving#structured-thinking

Sequential Thinking MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "sequential-thinking": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-sequential-thinking"
      ]
    }
  }
}

Sequential Thinking MCP Server 工具

1

A detailed tool for dynamic and reflective problem-solving through thoughts. This tool helps analyze problems through a flexible thinking process that can adapt and evolve. Each thought can build on, question, or revise previous insights as understanding deepens. When to use this tool: - Breaking down complex problems into steps - Planning and design with room for revision - Analysis that might need course correction - Problems where the full scope might not be clear initially - Problems that require a multi-step solution - Tasks that need to maintain context over multiple steps - Situations where irrelevant information needs to be filtered out Key features: - You can adjust total_thoughts up or down as you progress - You can question or revise previous thoughts - You can add more thoughts even after reaching what seemed like the end - You can express uncertainty and explore alternative approaches - Not every thought needs to build linearly - you can branch or backtrack - Generates a solution hypothesis - Verifies the hypothesis based on the Chain of Thought steps - Repeats the process until satisfied - Provides a correct answer Parameters explained: - thought: Your current thinking step, which can include: * Regular analytical steps * Revisions of previous thoughts * Questions about previous decisions * Realizations about needing more analysis * Changes in approach * Hypothesis generation * Hypothesis verification - nextThoughtNeeded: True if you need more thinking, even if at what seemed like the end - thoughtNumber: Current number in sequence (can go beyond initial total if needed) - totalThoughts: Current estimate of thoughts needed (can be adjusted up/down) - isRevision: A boolean indicating if this thought revises previous thinking - revisesThought: If is_revision is true, which thought number is being reconsidered - branchFromThought: If branching, which thought number is the branching point - branchId: Identifier for the current branch (if any) - needsMoreThoughts: If reaching end but realizing more thoughts needed You should: 1. Start with an initial estimate of needed thoughts, but be ready to adjust 2. Feel free to question or revise previous thoughts 3. Don't hesitate to add more thoughts if needed, even at the "end" 4. Express uncertainty when present 5. Mark thoughts that revise previous thinking or branch into new paths 6. Ignore information that is irrelevant to the current step 7. Generate a solution hypothesis when appropriate 8. Verify the hypothesis based on the Chain of Thought steps 9. Repeat the process until satisfied with the solution 10. Provide a single, ideally correct answer as the final output 11. Only set nextThoughtNeeded to false when truly done and a satisfactory answer is reached

概览

什么是 Sequential Thinking MCP Server?

Sequential Thinking MCP 服务器是 Model Context Protocol 项目推出的官方参考服务器,它为模型提供一个用于结构化、分步骤解决问题的工具,并允许修改先前的步骤、分支到其他推理路径。

模型不再一次性给出答案,而是记录带编号的思考步骤,随时调整预计所需的步数,标注修订,并在给出最终答案前探索不同分支。它在 npm 上以 @modelcontextprotocol/server-sequential-thinking 发布,Docker 镜像为 mcp/sequentialthinking。它不需要 API 密钥,也不调用任何外部服务。

如何使用 Sequential Thinking MCP Server?

前提条件:Node.js(用于 npx)或 Docker。无需设置环境变量,唯一可选项是 DISABLE_THOUGHT_LOGGING=true,用于禁止服务器记录思考内容。

Claude Desktop

在 claude_desktop_config.json 中添加:

{
  "mcpServers": {
    "sequential-thinking": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-sequential-thinking"
      ]
    }
  }
}

在 Windows 上,通过 cmd 启动 npx:

{
  "mcpServers": {
    "sequential-thinking": {
      "command": "cmd",
      "args": ["/c", "npx", "-y", "@modelcontextprotocol/server-sequential-thinking"]
    }
  }
}

使用 Docker 时,设置 "command": "docker" 和 "args": ["run", "--rm", "-i", "mcp/sequentialthinking"]。

Claude Code

claude mcp add sequential-thinking -- npx -y @modelcontextprotocol/server-sequential-thinking

Cursor

Cursor 使用相同的 mcpServers 格式。将上面的 npx 配置块添加到 ~/.cursor/mcp.json(全局)或 .cursor/mcp.json(项目)。

VS Code

在命令面板中运行 MCP: Open User Configuration,或在工作区中创建 .vscode/mcp.json:

{
  "servers": {
    "sequential-thinking": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-sequential-thinking"
      ]
    }
  }
}

Codex

codex mcp add sequential-thinking npx -y @modelcontextprotocol/server-sequential-thinking

添加服务器后,重启或重新加载客户端,确认工具列表中出现 sequential_thinking,然后让模型分步骤解决一个有难度的问题。

Sequential Thinking MCP Server 的主要功能

  • 将复杂问题拆解为带编号、易于处理的步骤
  • 随着理解加深,允许模型修改之前的思考
  • 支持使用分支 ID 分出其他推理路径
  • 在推理过程中增加或减少计划的思考步数
  • 鼓励先提出解决方案假设,再进行验证
  • 完全在本地运行,无需 API 密钥或外部服务
  • 可通过 npx 和 Docker 使用;采用 MIT 许可证

Sequential Thinking MCP Server 工具

该服务器只提供一个工具:sequential_thinking。每次调用承载一步推理:

  • 核心字段: thought(当前步骤)、thoughtNumber、totalThoughts(当前估计的总步数)和 nextThoughtNeeded(是否继续)。
  • 修订字段: isRevision 和 revisesThought 标记重新审视先前步骤的思考。
  • 分支字段: branchFromThought 和 branchId 用于开启并标记另一条路径。
  • 扩展字段: needsMoreThoughts 表示模型已到达原计划的终点,但还需要更多步骤。

何时调用该工具由宿主应用决定。除非客户端开放了原始工具调用,否则通常不需要手动调用。

Sequential Thinking MCP Server 的使用场景

  • 迁移规划:“Plan a database migration from PostgreSQL 14 to 16, list risks, and revise the plan if downtime exceeds 5 minutes.”
  • 调试:“Debug why this deployment only fails in production and show your reasoning step by step.”
  • 架构决策:“Compare three architecture options for a file sync engine and branch if one assumption turns out to be wrong.”
  • 起初范围不清晰、计划需要预留调整空间的规划与设计工作。
  • 需要跨步骤保持上下文并过滤无关信息的长流程多步骤任务。

Sequential Thinking MCP Server 与替代方案对比

Sequential Thinking 只在一次对话内组织推理,对话结束后不会记住任何内容。如果需要跨会话持久化,可以搭配同一仓库中的另一个参考服务器 Memory MCP Server,它会存储知识图谱。许多模型自带扩展推理模式;而 sequential thinking MCP 服务器与模型无关,并把每一步都以工具调用的形式呈现,便于检查或记录推理过程。社区中也有若干分支版本增加了 HTTP 传输等功能,但本页面收录的是官方参考实现。

Sequential Thinking MCP Server 常见问题排查

  • 工具没有出现: 重启或重新加载宿主应用使其重新连接,然后在 MCP 工具列表或调试器中查找 sequential_thinking。
  • 模型从不调用它: 针对有难度的问题明确要求分步骤推理。是否使用该工具由宿主决定。
  • Windows 上 npx 无法启动: 按上文使用 "command": "cmd" 并加上 "/c", "npx", ...。
  • 日志中出现思考内容: 将 DISABLE_THOUGHT_LOGGING 设为 true,服务器就不会记录思考信息。
  • 工具名与文档不一致: 当前版本的工具名为 sequential_thinking,部分旧的收录页面显示为 sequentialthinking。请更新到最新的 npm 包或 Docker 镜像。

常见问题

1

Sequential Thinking MCP 服务器是官方的吗?

是的。它位于 `modelcontextprotocol/servers` 仓库的 `src/sequentialthinking` 目录,由 Model Context Protocol 项目作为参考服务器维护。

2

它免费吗?需要 API 密钥吗?

免费,采用 MIT 许可证。由于不调用任何外部服务,因此不需要 API 密钥或账号。

3

sequential thinking MCP 支持 Claude Code、Cursor 和 VS Code 吗?

支持。它是标准的 stdio 服务器,任何 MCP 宿主都能运行。README 提供了 Claude Desktop、VS Code 和 Codex 的配置;Claude Code 和 Cursor 使用同样的 npx 命令。

4

它会让模型变得更聪明吗?

它不会增加知识,也不会改变模型本身。它为模型提供一个结构化的空间来拆解、修订和分支推理,这对一次性回答容易遗漏步骤的问题有帮助。

5

我的数据会被发送到别处吗?

不会。服务器在本地运行,只跟踪会话中传给它的思考内容。除非将 `DISABLE_THOUGHT_LOGGING` 设为 `true`,否则它可能会在本地记录思考内容。

6

可以用 Docker 运行吗?

可以。在客户端配置中将 `docker run --rm -i mcp/sequentialthinking` 作为启动命令。

7

如何判断它是否在正常工作?

宿主应用或调试器中应能看到多次 `sequential_thinking` 调用,带有 `thoughtNumber`、`totalThoughts` 和 `nextThoughtNeeded` 等字段,而不是一次性给出答案。

评论

其他 分类下的更多 MCP 服务器