Can MCP Server

🚀 Concept Activation Network (CAN) MCP Server

The Concept Activation Network (CAN) MCP Server implements a parallel associative thinking approach through the Model Context Protocol (MCP). Unlike traditional sequential reasoning methods, CAN uses an interconnected network of concepts where activation energy spreads in parallel until a coherent pattern or solution emerges.

🚀 Quick Start

The Concept Activation Network (CAN) MCP Server implements a parallel associative thinking approach through the Model Context Protocol (MCP). Different from traditional sequential reasoning methods, CAN uses an interconnected network of concepts, in which activation energy spreads in parallel until a coherent pattern or solution emerges.

✨ Features

Parallel Concept Activation

Traditional AI reasoning typically relies on sequential, step-by-step thinking processes. CAN adopts a fundamentally different approach:

1. Network Structure: Knowledge is represented as an interconnected network of concepts.
2. Parallel Activation: When stimulated, activation energy spreads simultaneously through multiple pathways.
3. Emergent Patterns: Solutions emerge as stable patterns formed by highly active related concepts.
4. Nonlinear Exploration: Multiple conceptual paths are explored simultaneously.

This approach is inspired by human cognitive theories, which suggest that we often think through associations rather than pure logic.

Core Components

The CAN system includes:

1. Concept Nodes: Basic units representing specific concepts, ideas, or elements.
2. Weighted Connections: Links between concepts with different strengths.
3. Activation Dynamics: Algorithms that control how activation energy spreads in the network.
4. Pattern Detection: Methods for identifying emergent structures.

📦 Installation

# Clone the repository
git clone https://github.com/psikosen/can-mcp-server.git
cd can-mcp-server

# Install dependencies
npm install

# Start the server
npm start

📚 Documentation

Configuration in Claude Desktop

To use the CAN MCP Server with Claude Desktop, add the following to claude_desktop_config.json:

{
  "mcpServers": {
    "can": {
      "command": "node",
      "args": ["yourpath/can-mcp-server/can-server.js"]
    }
  }
}

MCP Tools

The CAN MCP Server provides the following tools:

Session Management

• create_session: Create a new CAN session.
• get_all_sessions: List all active sessions.
• delete_session: Delete a session.

Concept Management

• add_concept: Add a concept to the network.
• remove_concept: Remove a concept from the network.
• add_connection: Create a connection between concepts.
• remove_connection: Remove a connection between concepts.

Activation Process

• set_parameters: Configure activation parameters.
• set_initial_activation: Set the initial activation value of a concept.
• activate: Spread activation energy in the network.
• deactivate: Stop the spread of activation.
• get_activation: Get the current activation state of a specified concept.

Pattern Detection

• detect_pattern: Identify significant patterns or structures in the network.
• query: Query the network based on specific conditions.
• monitor: Monitor changes in the network state.

💻 Usage Examples

Basic Usage

1. Create a session

create_session "My First CAN Session"

2. Add concepts and connections

add_concept "Artificial Intelligence"
add_concept "Machine Learning"
add_connection "Artificial Intelligence" "Machine Learning" 0.8

3. Spread activation and get results

set_initial_activation "Artificial Intelligence" 1.0
activate
get_activation "Machine Learning"

4. Delete a session

delete_session "My First CAN Session"

📚 Documentation

When to Use CAN

Recommended Scenarios

• Associative Reasoning: When dealing with associations between concepts.
• Pattern Recognition: When extracting meaningful structures from complex information.
• Creative Thinking: When a non - linear, divergent thinking process is required.
• Knowledge Graph Construction: When representing and managing complex knowledge networks.

Not Recommended Scenarios

• Logical Reasoning: For reasoning tasks that require strict adherence to logical rules.
• Mathematical Calculation: For tasks that require precise numerical calculations.
• Sequential Processing: For tasks that need to be executed strictly step - by - step.
• Simple Queries: For tasks that only require basic fact retrieval.

🤝 Contribution

Contributions to improve the CAN MCP Server are welcome! Please submit pull requests or open issues to discuss potential improvements.

📄 License

This project is open - sourced under the MIT License. For more details, please refer to the LICENSE file.