Context-Optimizer-MCP

MrNitro360/Context-Optimizer-MCP

3.1

If you are the rightful owner of Context-Optimizer-MCP and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The Context Optimizer MCP server is designed to enhance AI assistants' ability to manage conversation context efficiently, ensuring better continuity and reduced token usage.

Tools
5
Resources
0
Prompts
0

Context Optimizer MCP

A Model Context Protocol (MCP) server for intelligent conversation context management with token optimization, designed to enhance Claude's ability to retain and utilize conversation context across interactions.

🎯 Purpose

This MCP server provides advanced context management capabilities that help AI assistants maintain better conversation continuity, reduce token usage through intelligent optimization, and prevent context loss during long conversations.

✨ Features

Core Context Management

  • Context Storage & Retrieval: Store and retrieve important conversation context with categorization
  • Conversation Summarization: Generate intelligent summaries to preserve key information
  • Topic Flow Tracking: Monitor conversation evolution and topic transitions
  • Actionable Item Extraction: Identify and track tasks, decisions, and follow-ups

Token Optimization

  • Smart Compression: Multiple compression strategies to reduce token usage
  • Budget Management: Configurable token budgets with automatic optimization
  • Usage Analytics: Detailed token usage statistics and recommendations
  • Priority-Based Filtering: Preserve high-importance context while optimizing low-priority content

Advanced Features

  • Context Gap Detection: Identify missing context that could improve assistance quality
  • Anti-Hallucination Validation: Built-in validation to ensure context accuracy
  • Persistent Storage: Optional permanent storage for critical context
  • Flexible Configuration: Customizable settings for different use cases

🚀 Installation

Prerequisites

  • Node.js ≥ 18.0.0
  • Claude Desktop or compatible MCP client

Setup

  1. Clone the repository:
git clone https://github.com/MrNitro360/Context-Optimizer-MCP.git
cd Context-Optimizer-MCP
  1. Install dependencies:
npm install
  1. Configure Claude Desktop to use this MCP server by adding to your claude_desktop_config.json:
{
  "mcpServers": {
    "context-optimizer": {
      "command": "node",
      "args": ["path/to/Context-Optimizer-MCP/src/index.js"]
    }
  }
}
  1. Start the server:
npm start

🛠️ Available Tools

Context Management Tools

store_conversation_context

Store important conversation context for future reference.

  • Parameters: contextType, content, importance (1-10), tags, expiresAfter
  • Context Types: decision, preference, fact, goal, constraint, insight
retrieve_relevant_context

Retrieve stored context relevant to the current conversation.

  • Parameters: query, contextTypes, maxResults, minImportance, tokenBudget
summarize_conversation

Create intelligent summaries of conversation content.

  • Parameters: conversationText, focusAreas, includeCodeContext, tokenBudget
track_conversation_flow

Track conversation evolution and topic transitions.

  • Parameters: currentMessage, messageType, topicShift
extract_actionable_items

Extract tasks, decisions, and follow-ups from conversation.

  • Parameters: conversationText, itemTypes

Optimization Tools

optimize_context_for_continuation

Optimize and compress context for efficient conversation continuation.

  • Parameters: fullContext, targetLength, preserveTypes, tokenBudget
detect_context_gaps

Identify missing context that could improve assistance quality.

  • Parameters: currentRequest, availableContext, domain
get_token_usage_stats

Get detailed token usage statistics and optimization recommendations.

  • Parameters: includeRecommendations, analyzeRedundancy
optimize_tokens

Apply token optimization using various compression strategies.

  • Parameters: strategy, targetReduction, preserveImportant
  • Strategies: priority_filtering, content_compression, context_merging, aggressive_compression
set_token_budget

Configure token budgets and optimization settings.

  • Parameters: maxContextTokenBudget, compressionThreshold, tokenOptimizationEnabled, maxTokensPerContext

🔧 Configuration

Default Settings

  • Max Context Token Budget: 8,000 tokens
  • Compression Threshold: 0.8 (80%)
  • Max Tokens Per Context: 1,000 tokens
  • Token Optimization: Enabled

Environment Variables

Set these environment variables to customize behavior:

  • CONTEXT_STORAGE_PATH: Directory for persistent context storage
  • MAX_TOKEN_BUDGET: Override default token budget
  • DEBUG_MODE: Enable detailed logging

📖 Usage Examples

Basic Context Storage

// Store a user preference
await store_conversation_context({
  contextType: "preference",
  content: "User prefers concise explanations with code examples",
  importance: 8,
  tags: ["communication", "coding"],
  expiresAfter: "permanent"
});

Context Retrieval

// Retrieve relevant context for current question
await retrieve_relevant_context({
  query: "How should I explain this technical concept?",
  minImportance: 6,
  maxResults: 5
});

Token Optimization

// Optimize stored context to reduce token usage
await optimize_tokens({
  strategy: "priority_filtering",
  targetReduction: 25,
  preserveImportant: true
});

🏗️ Architecture

Core Components

  • ContextOptimizerServer: Main MCP server implementation
  • ContextOptimizer: Advanced context processing and optimization
  • TokenOptimizer: Token usage analysis and compression
  • FileManager: Persistent storage management
  • DateUtils: Time-based context management

Data Flow

  1. Context ingestion through MCP tools
  2. Processing and categorization
  3. Storage in memory/disk with metadata
  4. Retrieval with relevance scoring
  5. Token optimization when needed

🧪 Development

Scripts

  • npm start: Start the production server
  • npm run dev: Start with nodemon for development
  • npm run lint: Check code style
  • npm run format: Format code with Prettier

Testing

Run the server in test mode:

node src/index.js

🤝 Contributing

  1. Fork the repository
  2. Create a feature branch (git checkout -b feature/amazing-feature)
  3. Commit changes (git commit -m 'Add amazing feature')
  4. Push to branch (git push origin feature/amazing-feature)
  5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the file for details.

🔗 Links

📊 Performance

  • Memory Efficient: Intelligent context pruning and compression
  • Fast Retrieval: Optimized search algorithms for context lookup
  • Scalable: Handles thousands of context items efficiently
  • Token Optimized: Reduces token usage by up to 50% through compression

Built with ❤️ for the Claude AI ecosystem by mrnitro360