Sequential Thinking Mcp Server

Cloud MCP Server for structured problem-solving. AI agents break down complex problems into thinking sequences with revisions, branching, and adaptive analysis.

Features

• Adaptive Totals - Adjust estimated thoughts up/down as you progress
• Revisions - Question or revise any previous thought
• Branching - Explore alternative approaches with named branches
• Session Persistence - Resume thinking sessions across runs
• Cloud Deployment - No local setup required

How It Works

The server tracks a chain of thoughts. Each thought can:

1. Continue the sequence (standard thought)
2. Revise a previous thought (correction)
3. Branch into an alternative path (exploration)

Sessions persist to Apify datasets, enabling resume and review.

Input Schema

Parameter	Type	Required	Default	Description
thought	string	Yes	-	Your current thinking step
thoughtNumber	integer	No	1	Position in sequence
totalThoughts	integer	No	5	Estimated total needed
nextThoughtNeeded	boolean	No	true	Continue thinking?
isRevision	boolean	No	false	Revising previous?
revisesThought	integer	No	-	Thought being revised
branchFromThought	integer	No	-	Branch point
branchId	string	No	-	Branch identifier
sessionId	string	No	-	Resume session
persistSession	boolean	No	true	Save to dataset

Output Schema

Field	Type	Description
sessionId	string	Session identifier
thoughtNumber	integer	Current position
totalThoughts	integer	Estimated total
nextThoughtNeeded	boolean	Continue flag
branches	array	All branch IDs
thoughtHistoryLength	integer	Thoughts recorded
status	string	success or error

Examples

Basic Thinking Step

{
  "thought": "First, identify the core system components...",
  "thoughtNumber": 1,
  "totalThoughts": 5,
  "nextThoughtNeeded": true
}

Revision

{
  "thought": "Component B should precede component A...",
  "thoughtNumber": 3,
  "totalThoughts": 5,
  "isRevision": true,
  "revisesThought": 2,
  "nextThoughtNeeded": true
}

Branching

{
  "thought": "Exploring microservices alternative...",
  "thoughtNumber": 4,
  "totalThoughts": 6,
  "branchFromThought": 2,
  "branchId": "microservices-approach",
  "nextThoughtNeeded": true
}

Resume Session

{
  "thought": "Continuing analysis...",
  "sessionId": "session_1703123456789_abc123",
  "thoughtNumber": 6,
  "totalThoughts": 8,
  "nextThoughtNeeded": true
}

MCP Integration

Claude Desktop Configuration

{
  "mcpServers": {
    "sequential-thinking": {
      "command": "npx",
      "args": ["-y", "@anthropic/apify-mcp-client", "constant_quadruped/sequential-thinking-mcp-server"]
    }
  }
}

Apify Client (JavaScript)

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_TOKEN' });

const run = await client.actor('constant_quadruped/sequential-thinking-mcp-server').call({
  thought: "Analyzing step by step...",
  thoughtNumber: 1,
  totalThoughts: 5,
  nextThoughtNeeded: true
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Apify Client (Python)

from apify_client import ApifyClient

client = ApifyClient("YOUR_TOKEN")

run = client.actor("constant_quadruped/sequential-thinking-mcp-server").call(run_input={
    "thought": "Analyzing step by step...",
    "thoughtNumber": 1,
    "totalThoughts": 5,
    "nextThoughtNeeded": True
})

items = client.dataset(run["defaultDatasetId"]).list_items().items
print(items)

Use Cases

Software Architecture

Break down system design into components, evaluate trade-offs, revise based on constraints.

Thought 1: Identify core requirements
Thought 2: Propose initial architecture
Thought 3: (Revision) Adjust for scalability
Thought 4: (Branch) Explore serverless
Thought 5: Compare and recommend

Research Analysis

Analyze data step-by-step, branch into hypotheses, revise as patterns emerge.

Debugging

Trace issues, branch into potential causes, eliminate hypotheses systematically.

Strategic Planning

Structure thinking, explore alternatives, refine based on constraints.

Session Persistence

Sessions save to Apify datasets automatically:

• Resume: Pass sessionId to continue
• Review: Access full thought chain
• Compare: Analyze different branches
• Export: Download complete sessions

Resources

• Model Context Protocol
• MCP Servers Repository
• Apify Documentation

License

MIT - Based on @modelcontextprotocol/server-sequential-thinking