Chat Analysis

🚀 MCP Chat Analysis Server

A server based on the Model Context Protocol (MCP) that enables semantic analysis of chat conversations through vector embeddings and knowledge graphs. The server provides tools for analyzing chat data, performing semantic searches, extracting concepts, and analyzing conversation patterns.

🚀 Quick Start

# Install the package
pip install mcp-chat-analysis-server

# Set up the configuration
cp config.example.yml config.yml
# Edit the database settings using config.yml

# Run the server
python -m mcp_chat_analysis.server

✨ Features

• 🔍 Semantic Search: Find relevant messages and conversations using vector similarity.
• 🕸️ Knowledge Graph: Navigate relationships between messages, concepts, and topics.
• 📊 Conversation Analysis: Analyze patterns, metrics, and conversation dynamics.
• 🔄 Flexible Import: Support multiple chat export formats.
• 🚀 MCP Integration: Easily integrate with Claude and other MCP-compatible systems.

📦 Installation

1. Install the package:

pip install mcp-chat-analysis-server

2. Set up the database:

# Use Docker (recommended)
docker compose up -d

3. Configure the server:

cp .env.example .env
# Edit your settings using .env

💻 Usage Examples

MCP Integration

Add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "chat-analysis": {
      "command": "python",
      "args": ["-m", "mcp_chat_analysis.server"],
      "env": {
        "QDRANT_URL": "http://localhost:6333",
        "NEO4J_URL": "bolt://localhost:7687",
        "NEO4J_USER": "neo4j",
        "NEO4J_PASSWORD": "Your password"
      }
    }
  }
}

Available Tools

Basic Usage

import_conversations

Import and analyze chat conversations.

{
    "source_path": "/path/to/export.zip",
    "format": "openai_native"  # or html, markdown, json
}

semantic_search

Search conversations by semantic similarity.

{
    "query": "Machine learning applications",
    "limit": 10,
    "min_score": 0.7
}

analyze_metrics

Analyze conversation metrics.

{
    "conversation_id": "conv-123",
    "metrics": [
        "message_frequency",
        "response_times",
        "topic_diversity"
    ]
}

extract_concepts

Extract and analyze concepts.

{
    "conversation_id": "conv-123",
    "min_relevance": 0.5,
    "max_concepts": 10
}

📚 Documentation

Architecture

Refer to ARCHITECTURE.md for detailed diagrams and documentation to understand:

• System components and interactions
• Data flow and processing procedures
• Storage patterns and vector operations
• Tool integration mechanisms

Prerequisites

• Python 3.8+
• Neo4j database for knowledge graph storage
• Qdrant vector database for semantic search
• sentence-transformers for embeddings

Development

1. Clone the repository:

git clone https://github.com/rebots-online/mcp-chat-analysis-server.git
cd mcp-chat-analysis-server

2. Install development dependencies:

pip install -e ".[dev]"

3. Run tests:

pytest tests/

Contribution

1. Fork the repository.
2. Create a feature branch.
3. Submit a pull request.

Refer to CONTRIBUTING.md for guidelines.

License

MIT License - See the LICENSE file for details.

Related Projects

• Model Context Protocol (MCP)
• Claude Desktop
• Qdrant Vector Database
• Neo4j Graph Database

Support

• 📖 Documentation
• 🐛 Issue Tracker
• 💬 Discussion Area