Thinking Protocol MCP Server

A Model Context Protocol (MCP) server that implements a structured 6-stage thinking framework for systematic problem-solving and decision-making.

🎯 Overview

This MCP server provides tools to guide AI agents through a rigorous thinking process with 6 mandatory stages:

1. OBSERVE - Capture objective inputs only
2. ANALYZE - Derive meaning and context
3. ROOT - Identify true underlying cause
4. ACT - Define actionable decisions
5. VALIDATE - Verify outcome and correctness
6. IMPROVE - Prevent recurrence and strengthen system

🚀 Installation

Option 1: Via NPM (Recommended - After Publishing)

npm install -g thinking-protocol-mcp

Option 2: From GitHub

git clone https://github.com/RimGit-N/thinking-protocol-mcp.git
cd thinking-protocol-mcp
npm install

Option 3: Via NPX (No Installation Required)

npx -y thinking-protocol-mcp

🔧 Configuration

Add to your MCP configuration file (e.g., mcp_config.json or Claude Desktop config):

Using NPX (Recommended)

{
  "mcpServers": {
    "thinking-protocol": {
      "command": "npx",
      "args": [
        "-y",
        "thinking-protocol-mcp"
      ]
    }
  }
}

Using Global Installation

{
  "mcpServers": {
    "thinking-protocol": {
      "command": "thinking-protocol-mcp",
      "args": []
    }
  }
}

Using Local Clone

{
  "mcpServers": {
    "thinking-protocol": {
      "command": "node",
      "args": [
        "/absolute/path/to/thinking-protocol-mcp/index.js"
      ]
    }
  }
}

Complete Example with Multiple MCP Servers

If you're using multiple MCP servers together:

{
  "mcpServers": {
    "chrome-devtools": {
      "command": "npx",
      "args": [
        "-y",
        "chrome-devtools-mcp"
      ]
    },
    "context7": {
      "command": "npx",
      "args": [
        "-y",
        "@upstash/context7-mcp"
      ]
    },
    "thinking-protocol": {
      "command": "npx",
      "args": [
        "-y",
        "thinking-protocol-mcp"
      ]
    }
  }
}

Note: Make sure there's a comma (,) between each server entry, except after the last one.

🛠️ Available Tools

1. thinking_start_session

Start a new thinking protocol session.

Input:

{
  "task_description": "Description of the problem to solve"
}

Output:

{
  "success": true,
  "session_id": "session_1234567890_abc123",
  "next_stage": "observe",
  "protocol": { ... }
}

2. thinking_execute_stage

Execute a specific stage of the protocol.

Input:

{
  "session_id": "session_1234567890_abc123",
  "stage": "observe",
  "data": {
    "facts": "Factual data...",
    "signals": "Important indicators...",
    "state": "Current system state..."
  }
}

Stage Requirements:

Stage	Required Keys
observe	facts, signals, state
analyze	patterns, assumptions, context
root	primary_cause, supporting_factors, core_issue
act	main_action, alternatives_risks, execution_priority
validate	success_metrics, failure_signs, evaluation_method
improve	system_change, rules_sop, long_term_prevention

3. thinking_get_status

Get current status of a thinking session.

Input:

{
  "session_id": "session_1234567890_abc123"
}

4. thinking_get_protocol

Get the complete protocol configuration.

Input:

{}

5. thinking_validate_session

Validate if all stages are completed correctly.

Input:

{
  "session_id": "session_1234567890_abc123"
}

📋 Usage Example

// 1. Start session
const session = await thinking_start_session({
  task_description: "Optimize database query performance"
});

// 2. Execute OBSERVE stage
await thinking_execute_stage({
  session_id: session.session_id,
  stage: "observe",
  data: {
    facts: "Query takes 5 seconds, 10M rows, no indexes",
    signals: "CPU usage spikes to 100% during query",
    state: "Production database, peak hours"
  }
});

// 3. Execute ANALYZE stage
await thinking_execute_stage({
  session_id: session.session_id,
  stage: "analyze",
  data: {
    patterns: "Full table scans on large dataset",
    assumptions: "No proper indexing strategy",
    context: "Legacy system, no recent optimization"
  }
});

// ... continue with remaining stages

// 6. Validate session
const validation = await thinking_validate_session({
  session_id: session.session_id
});

🔒 Execution Rules

• Mode: STRICT - All stages must be completed
• Ordered: TRUE - Stages must be executed in sequence
• Allow Skip: FALSE - Cannot skip any stage
• Require All Keys: TRUE - All required keys must be filled

🎓 When to Use

Mandatory for:

• ✅ Complex problem-solving
• ✅ Root cause analysis
• ✅ Architectural decisions
• ✅ Security analysis
• ✅ Performance optimization

Optional for:

• ⚪ Simple, straightforward tasks
• ⚪ Routine operations

📄 License

MIT

👤 Author

Rimuru