Critical Thinking MCP Server

A dual-perspective thinking analysis server based on Model Context Protocol (MCP), providing comprehensive performance evaluation through Actor-Critic methodology integrated with Critical Thinking best practices.

How to use

• build firstly with npm install && npm run build

  • It will generate dist/index.js file

• config this mcp in your mcp client such as Claude Code

{
  "mcpServers": {
    "critical-thinking": {
      "command": "node",
      "args": [
        "<path>/critical-thinking-mcp/dist/index.js"
      ],
      "env": {}
    }
  }
}

Critical Thinking Framework

This server implements a dialectical process using two core components of critical thinking:

1. Elements of Thought (The 'Actor's Guide')

The Actor (performer/creator) uses these elements to articulate their position:

• Purpose: What is the goal of this action or decision?
• Question at Issue: What specific problem is being addressed?
• Information/Evidence: What data or facts are being used?
• Assumptions: What is being taken for granted?
• Concepts: What theories or definitions govern this thinking?
• Inferences: What interpretations are being made?
• Implications: What happens if this line of thought is followed?
• Points of View: From what perspective are we looking at this?

2. Universal Intellectual Standards (The 'Critic's Checklist')

The Critic (evaluator/analyst) evaluates the Actor's input against these standards:

• Clarity: Is the point stated clearly?
• Accuracy: Is the claim true and verifiable?
• Precision: Are enough specific details provided?
• Relevance: How does this relate to the core problem?
• Depth: Does it address the complexities?
• Breadth: Are other perspectives considered?
• Logic: Does the conclusion follow from the premises?
• Significance: Is this the most important factor?
• Fairness: Is the assessment unbiased and empathetic?

Parameters

Required Parameters

• content (string): Current analysis content from the specified role perspective.
• role (string): Perspective role: "actor" (creative/empathetic) or "critic" (analytical/evaluative).
• nextRoundNeeded (boolean): Whether another round of actor-critic dialogue is needed.
• thoughtNumber (integer): Current thought number in the sequence (starts at 1).
• totalThoughts (integer): Total planned thoughts (must be odd and >= 3).

Optional Parameters (Critical Thinking Metadata)

• assumptions (string[]): Key assumptions being made in this thought.
• evidence (string[]): Data, facts, or observations supporting this thought.
• standardsApplied (string[]): Intellectual standards (e.g., Clarity, Logic) applied (primarily for Critics).

Best Practices

1. Dialectical Iteration: Always alternate between Actor and Critic roles to foster a self-correcting dialogue.
2. Explicit Evidence: Use the evidence and assumptions parameters to make the underlying logic transparent.
3. Apply Standards: Critics should explicitly state which Intellectual Standards they are using to evaluate the Actor's work.
4. Depth over Speed: Don't rush to set nextRoundNeeded to false. Ensure the analysis has reached a robust, multi-dimensional consensus.
5. Odd Thought Totals: Always plan for an odd number of total thoughts to ensure a balanced sequence (e.g., Actor -> Critic -> Actor/Final Synthesis).