sequential-thinking

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Sequential Thinking

A tool for dynamic, reflective problem-solving through a chain of numbered thoughts. Full parity with the Sequential Thinking MCP server — same parameters, same state management, same behavioral contract.

一种通过一系列编号思考实现动态、反思性问题解决的工具。与Sequential Thinking MCP服务器完全对等——相同的参数、相同的状态管理、相同的行为约定。

How to Use This Skill

如何使用该Skill

When this skill is activated, use
scripts/think.ts
as your primary reasoning mechanism. Do not reason in prose — reason through the script. Every step of your analysis should be a thought submitted via the script, making your reasoning chain explicit and trackable.

当激活此Skill时，请将
scripts/think.ts
作为主要推理机制。不要用散文形式推理——通过脚本进行推理。分析的每一步都应通过脚本提交为一个思考条目，让推理链清晰可追踪。

Workflow

工作流程

Reset state at the start of every new thinking session
Loop: Submit thoughts one at a time via the script, incrementing
```
thoughtNumber
```
each time
Adapt: Revise earlier thoughts, branch into alternatives, or extend depth as needed
Terminate: Set
```
nextThoughtNeeded false
```
only when you have a confident final answer
Respond: After the final thought, provide the answer to the user

Each thought should be a single Bash tool call. Think in the thought, not outside it.

重置：在每个新思考会话开始时重置状态
循环：通过脚本逐个提交思考条目，每次递增
```
thoughtNumber
```
调整：根据需要修订之前的思考、分支到替代方案或扩展深度
终止：仅当你有确定的最终答案时，设置
```
nextThoughtNeeded false
```
响应：完成最终思考后，向用户提供答案

每个思考条目都应为单个Bash工具调用。在思考条目中进行思考，而非外部。

Script Location

脚本位置

scripts/think.ts

Run via

bun

from the skill's base directory.

scripts/think.ts

从Skill的基础目录通过

bun

运行。

Commands

命令

Reset (required before every new session)

重置（每个新会话前必须执行）

bash

bun scripts/think.ts --reset

bash

bun scripts/think.ts --reset

Submit a Thought

提交思考条目

Required flags:

--thought

--thoughtNumber

--totalThoughts

--nextThoughtNeeded

bash

bun scripts/think.ts \
  --thought "Your analysis for this step" \
  --thoughtNumber 1 \
  --totalThoughts 5 \
  --nextThoughtNeeded true

必填参数：

--thought

、

--thoughtNumber

、

--totalThoughts

、

--nextThoughtNeeded

bash

bun scripts/think.ts \
  --thought "此步骤的分析内容" \
  --thoughtNumber 1 \
  --totalThoughts 5 \
  --nextThoughtNeeded true

Submit a Revision

提交修订条目

Revises a previous thought. The original stays in history; the revision is appended as a new entry. Requires

--isRevision

and

--revisesThought

bash

bun scripts/think.ts \
  --thought "Corrected analysis" \
  --thoughtNumber 3 \
  --totalThoughts 5 \
  --nextThoughtNeeded true \
  --isRevision --revisesThought 1

修订之前的思考条目。原始条目保留在历史记录中；修订内容作为新条目追加。需要

--isRevision

和

--revisesThought

参数。

bash

bun scripts/think.ts \
  --thought "修正后的分析内容" \
  --thoughtNumber 3 \
  --totalThoughts 5 \
  --nextThoughtNeeded true \
  --isRevision --revisesThought 1

Submit a Branch

提交分支条目

Explores an alternative path from a prior thought. Requires both

--branchFromThought

and

--branchId

bash

bun scripts/think.ts \
  --thought "Alternative approach" \
  --thoughtNumber 4 \
  --totalThoughts 7 \
  --nextThoughtNeeded true \
  --branchFromThought 2 --branchId alt-approach

探索从之前某个思考条目出发的替代路径。需要同时使用

--branchFromThought

和

--branchId

参数。

bash

bun scripts/think.ts \
  --thought "替代方法" \
  --thoughtNumber 4 \
  --totalThoughts 7 \
  --nextThoughtNeeded true \
  --branchFromThought 2 --branchId alt-approach

Extend Depth

扩展深度

Signal that more thoughts are needed beyond the original estimate.

bash

bun scripts/think.ts \
  --thought "Scope is larger than expected" \
  --thoughtNumber 6 \
  --totalThoughts 8 \
  --nextThoughtNeeded true \
  --needsMoreThoughts

表明需要超出最初预估的更多思考条目。

bash

bun scripts/think.ts \
  --thought "范围比预期更大" \
  --thoughtNumber 6 \
  --totalThoughts 8 \
  --nextThoughtNeeded true \
  --needsMoreThoughts

Inspect Full State

查看完整状态

bash

bun scripts/think.ts --status

Returns JSON with

fullHistory

and

branchDetails

bash

bun scripts/think.ts --status

返回包含

fullHistory

和

branchDetails

的JSON数据。

Output Format

输出格式

Each thought invocation prints the thought to stderr and a compact status line to stdout:

💭 Thought 3/7
The analysis shows that...
[3/7] history=3 next=true

Revision and branch thoughts use

🔄 Revision

and

🌿 Branch

headers respectively.

每次思考调用会将思考内容打印到stderr，并将简洁的状态行打印到stdout：

💭 Thought 3/7
分析表明...
[3/7] history=3 next=true

修订和分支思考条目分别使用

🔄 Revision

和

🌿 Branch

标题。

Parameters (Full MCP Parity)

参数（与MCP完全对等）

Parameter	Type	Required	Description
`--thought`	string	yes	The content of this thinking step
`--thoughtNumber`	int >= 1	yes	Current thought number in the sequence
`--totalThoughts`	int >= 1	yes	Estimated total thoughts needed (adjustable)
`--nextThoughtNeeded`	bool	yes	`true` to continue, `false` to terminate
`--isRevision`	flag	no	Marks this thought as revising a previous one
`--revisesThought`	int >= 1	no	Which thought number is being revised (required with `--isRevision` )
`--branchFromThought`	int >= 1	no	Create a branch starting from this thought number
`--branchId`	string	no	Label for the branch (required with `--branchFromThought` )
`--needsMoreThoughts`	flag	no	Signal that `totalThoughts` should be expanded

参数	类型	是否必填	描述
`--thought`	string	是	此思考步骤的内容
`--thoughtNumber`	int >= 1	是	序列中的当前思考编号
`--totalThoughts`	int >= 1	是	预估所需的总思考条目数（可调整）
`--nextThoughtNeeded`	bool	是	`true` 表示继续， `false` 表示终止
`--isRevision`	标志	否	将此思考标记为对之前条目的修订
`--revisesThought`	int >= 1	否	被修订的思考编号（与 `--isRevision` 一起使用时必填）
`--branchFromThought`	int >= 1	否	创建从该思考编号开始的分支
`--branchId`	string	否	分支的标签（与 `--branchFromThought` 一起使用时必填）
`--needsMoreThoughts`	标志	否	表示需要扩展 `totalThoughts`

Behavioral Rules

行为规则

These match the MCP server's tool description exactly:

Start with an initial estimate of
```
totalThoughts
```
, but adjust freely as understanding deepens
Auto-adjust: If
```
thoughtNumber
```
exceeds
```
totalThoughts
```
, the script raises
```
totalThoughts
```
to match
Revise previous thoughts when you realize an earlier step was wrong or incomplete — set
```
--isRevision
```
and
```
--revisesThought N
```
. The original stays in history; the revision is appended
Branch to explore alternative reasoning paths — set both
```
--branchFromThought N
```
and
```
--branchId label
```
. This does not abandon the main line
Extend beyond the initial estimate at any time with
```
--needsMoreThoughts
```
and an increased
```
--totalThoughts
```
Express uncertainty — not every thought needs confidence. Questioning and exploring is encouraged
Filter noise — ignore information irrelevant to the current step
Generate hypotheses when you have enough evidence, then verify them against prior thoughts in the chain
Iterate hypothesis-verification cycles until satisfied with the answer
Terminate only when you have a satisfactory answer — set
```
--nextThoughtNeeded false
```
Non-linear paths are first-class — branching, backtracking, and revision are features, not failures

这些规则与MCP服务器的工具描述完全一致：

开始时预估
```
totalThoughts
```
，但可随着理解加深自由调整
自动调整：如果
```
thoughtNumber
```
超过
```
totalThoughts
```
，脚本会将
```
totalThoughts
```
提升至匹配值
修订：当你意识到之前的步骤有误或不完整时，修订之前的思考条目——设置
```
--isRevision
```
和
```
--revisesThought N
```
。原始条目保留在历史记录中；修订内容追加到记录中
分支：探索替代推理路径——同时设置
```
--branchFromThought N
```
和
```
--branchId label
```
。这不会放弃主线思考
扩展：随时使用
```
--needsMoreThoughts
```
并增加
```
--totalThoughts
```
，超出最初预估
表达不确定性——并非每个思考条目都需要有信心。鼓励提出疑问和探索
过滤干扰——忽略与当前步骤无关的信息
生成假设：当你有足够证据时生成假设，然后对照推理链中的之前条目进行验证
迭代：进行假设-验证循环，直到对答案满意为止
终止：仅当你得到满意答案时终止——设置
```
--nextThoughtNeeded false
```
非线性路径是一等公民——分支、回溯和修订是功能，而非失败

State Management

状态管理

State is persisted to
```
scripts/.think_state.json
```
between invocations
```
thoughtHistory[]
```
is append-only — thoughts are never deleted
```
branches{}
```
maps branch IDs to their thought arrays
```
--reset
```
clears all state for a fresh session
```
--status
```
dumps the full state as JSON for inspection

状态在调用之间持久化到
```
scripts/.think_state.json
```
```
thoughtHistory[]
```
是仅追加的——思考条目永远不会被删除
```
branches{}
```
将分支ID映射到对应的思考数组
```
--reset
```
清除所有状态以开始新会话
```
--status
```
以JSON格式导出完整状态供查看

Example Session

示例会话

See references/example-session.md for a complete worked example demonstrating normal thoughts, revisions, branches, and dynamic depth adjustment.

查看references/example-session.md获取完整的示例会话，展示常规思考、修订、分支和动态深度调整的使用。