MCP Context Provider

# Changelog All notable changes to the MCP Context Provider project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). ## [1.8.2] - 2025-09-18 ### Added - **Phase 3 Intelligent Learning Context Enhancements** - Auto-trigger patterns for automatic memory storage in all 8 context files - Session initialization with memory searches and startup notifications - Tool-specific pattern matching for intelligent automation - New `general_preferences_context.json` for user workflow tracking - Proactive context suggestions and effectiveness analysis capabilities ### Enhanced Context Files - `applescript_context.json`: AppleScript detection and automation triggers - `terraform_context.json`: IaC code and command pattern triggers - `azure_context.json`: Cloud resource and Azure CLI triggers - `git_context.json`: Version control workflow triggers - `date_awareness_context.json`: Temporal event and scheduling triggers - `dokuwiki_context.json`: Wiki syntax and documentation triggers - `general_preferences_context.json`: User preference and workflow triggers (NEW) ### Technical Improvements - Standardized session_initialization format across all context files - Comprehensive pattern-based automation framework - Confidence thresholds for trigger accuracy - Memory integration with search and storage actions - Tool-specific tagging for organized memory storage ### Fixed - Corrected session_initialization structure to match validation requirements ## [1.8.1] - 2025-09-18 ### Added - ### Changed - ### Fixed - ## [1.8.0] - 2025-09-18 ### Added - **Complete Build Automation System**: Comprehensive CI/CD pipeline and quality assurance - **GitHub Actions**: Automated release workflow with package building and GitHub releases on git tags - **Continuous Integration**: Multi-platform testing (Ubuntu, macOS, Windows) with Python 3.8-3.11 support - **Pre-commit Hooks**: Automatic validation with JSON formatting, context validation, and build testing - **Security Integration**: Automated secret scanning and sensitive file detection - **Single Source of Truth Architecture**: Eliminated duplication and maintenance issues - **Dynamic Build System**: `scripts/build_dxt.py` builds packages from single source automatically - **Automated Validation**: Comprehensive validation scripts for context files, build process, and versions - **Version Management**: `scripts/bump_version.py` for automated semantic versioning across all files - **Quality Assurance Framework**: Multi-layer validation and testing - **Context Validation**: `scripts/validate_contexts.py` validates JSON structure and required fields - **Build Testing**: `scripts/test_build.py` tests build process without creating packages - **Version Consistency**: `scripts/check_versions.py` ensures version synchronization across files ### Changed - **Repository Structure**: Removed duplicate DXT directory from version control - **Build Artifacts**: DXT directory now generated during build, not version controlled - **Clean Repository**: Build artifacts excluded via .gitignore for cleaner development - **Dynamic Context Loading**: Server discovers all 8 context files automatically - **Installation Process**: Updated to use automated build system - **Fresh Package Builds**: Installation script builds packages from latest source - **No Download Dependencies**: Eliminates version drift from pre-built packages - **Consistent Installation**: Always installs latest server with dynamic loading ### Fixed - **Context File Discovery**: Server now loads all context files instead of hardcoded 5 - **Version Synchronization**: Automated version management prevents inconsistencies - **Build Reproducibility**: Eliminates manual file copying and synchronization issues ## [1.7.0] - 2025-09-18 ### Changed - **Repository Structure**: Major reorganization for improved maintainability - Created `docs/` directory with `guides/` and `phase3/` subdirectories for all documentation - Created `scripts/` directory for installation and utility scripts - Created `tests/` directory for all test files - Moved all documentation files to appropriate subdirectories - Cleaned up root directory from 37 to 22 items for better navigation ### Fixed - Updated all documentation references to reflect new directory structure - Fixed installation script paths in README.md and CLAUDE.md - Corrected test file paths in documentation ### Removed - Outdated DXT package file (mcp-context-provider-1.1.0.dxt) ## [1.6.0] - 2025-09-17 ### Added - **Phase 3: Synergistic Integration with Intelligent Learning**: Complete intelligent context evolution system - `ContextLearningEngine`: Advanced pattern recognition and context optimization engine - Context effectiveness analysis with usage pattern tracking and scoring algorithms - Intelligent optimization suggestions based on real memory service data analysis - Automatic context optimization with pattern improvement, preference tuning, and rule refinement - Proactive context suggestions based on usage patterns and missing tool contexts - Session pattern learning for continuous improvement of initialization performance - **Real Memory Service Integration**: Full integration with mcp-memory-service - `MemoryServiceIntegration` class replacing simulation layer with actual memory service calls - Automatic storage of context changes, optimizations, and learning insights in persistent memory - Memory-driven effectiveness analysis using historical usage data - Context evolution tracking with comprehensive change history in memory service - **Advanced MCP Tools for Learning**: 4 new intelligent tools for context management - `analyze_context_effectiveness`: Analyze context effectiveness with memory-driven insights - `suggest_context_optimizations`: Generate global optimization suggestions based on usage patterns - `get_proactive_suggestions`: Provide proactive context suggestions for workflow improvement - `auto_optimize_context`: Automatically optimize contexts based on learning engine recommendations - **Intelligent Context Evolution**: Self-improving context system - Automatic detection of context usage patterns and effectiveness metrics - Learning-driven context rule refinement and pattern optimization - Proactive suggestion system for missing tool contexts and workflow improvements - Memory-backed insights for continuous context enhancement ### Enhanced - **ContextProvider Architecture**: Extended with full learning capabilities - Learning engine integration with session-aware pattern recognition - Automatic learning during session initialization for performance optimization - Memory service integration for persistent learning data storage - Context optimization framework with backup and validation systems ### Technical Implementation - **Learning Engine**: Sophisticated pattern recognition with effectiveness scoring - **Memory Integration**: Real mcp-memory-service with sqlite_vec backend for persistent learning - **Auto-Optimization**: Intelligent context modification with pattern improvement and preference tuning - **Proactive Intelligence**: Usage-based suggestions for missing contexts and workflow enhancements - **Session Learning**: Continuous improvement through session performance analysis ### Testing & Quality Assurance - **Comprehensive Test Suite**: `test_phase3_learning.py` with 7 test categories - Learning engine initialization and memory service integration testing - Context effectiveness analysis with real usage data simulation - Optimization suggestion generation with global pattern analysis - Session pattern learning with performance insight generation - Proactive suggestion system with missing context detection - Auto-optimization testing with preference tuning and rule refinement - Memory integration testing with real mcp-memory-service storage and retrieval - **100% Test Pass Rate**: All Phase 3 synergistic integration features fully verified - **Real Memory Service**: Comprehensive testing with actual mcp-memory-service backend ### Phase 3 Implementation Complete This release represents the completion of the full 3-phase roadmap: - **Phase 1**: Session initialization with memory service integration ✅ - **Phase 2**: Dynamic context file creation and management ✅ - **Phase 3**: Synergistic integration with intelligent learning ✅ The MCP Context Provider now offers: - **Intelligent Context Evolution**: Self-improving contexts that learn from usage patterns - **Memory-Driven Insights**: Persistent learning data stored in real memory service - **Automatic Optimization**: Context refinement based on effectiveness analysis - **Proactive Intelligence**: Suggestions for workflow improvements and missing contexts - **Enterprise Ready**: Complete learning framework for team knowledge propagation ## [1.5.0] - 2025-09-17 ### Added - **Dynamic Context Management System**: Complete runtime context file creation and modification - `create_context_file`: Create new context files dynamically with full validation - `update_context_rules`: Update existing context rules with backup and validation - `add_context_pattern`: Add patterns to auto-trigger sections for memory integration - Support for all context file sections: syntax_rules, preferences, auto_corrections, session_initialization - **Advanced Security Framework**: Multi-layer validation and protection - Context name validation: alphanumeric, underscore, hyphen only with length limits - Reserved name protection (system, admin, config, server) - JSON structure validation with type checking and required field enforcement - Input sanitization and comprehensive error reporting - Security-first approach with detailed validation feedback - **Automated Backup & Versioning System**: Complete change tracking and recovery - Timestamped backups created automatically before any context modifications - Organized backup storage in `contexts/backups/` directory - Version tracking with metadata updates and modification timestamps - Rollback capability through preserved backup files - **Memory Service Integration Ready**: Framework for intelligent context evolution - Context usage tracking framework for effectiveness analysis - Change history tracking for pattern recognition - Foundation for storing pattern effectiveness data in mcp-memory-service - Smart context-aware memory integration preparation ### Enhanced - **ContextProvider Architecture**: Extended with comprehensive dynamic management - Added validation methods: `_validate_context_name()`, `_validate_context_data()` - Implemented backup system: `_backup_context_file()` - Dynamic context operations: create, update, and pattern management - Enhanced error handling with detailed feedback and recovery options ### Technical Implementation - **Security-First Design**: Multiple validation layers prevent malformed or malicious contexts - **Atomic Operations**: All context modifications are atomic with backup-first approach - **Performance Optimized**: Efficient file operations with minimal I/O overhead - **Memory Integration Ready**: Prepared for seamless mcp-memory-service connection - **Extensible Framework**: Clean API design for easy addition of new management features ### Testing & Quality Assurance - **Comprehensive Test Suite**: `test_dynamic_context.py` with 6 test categories - Context name validation testing with edge cases - Context creation testing with success and failure scenarios - Context update testing with backup verification - Pattern addition testing for auto-trigger management - Security validation testing with malformed inputs - Backup system testing with versioning verification - **100% Test Pass Rate**: All dynamic context management features verified - **Security Testing**: Comprehensive validation of input sanitization and error handling ### Foundation for Phase 3 This release establishes the infrastructure for: - **Intelligent Context Evolution**: Learning from usage patterns and effectiveness - **Team Knowledge Propagation**: Shared context evolution across team members - **Advanced Pattern Recognition**: Automatic context optimization based on user behavior - **Enterprise Context Analytics**: Usage tracking and optimization recommendations ## [1.4.0] - 2025-09-17 ### Added - **Session Initialization System**: Implemented automatic context rule execution at session startup - New `execute_session_initialization()` method processes `session_initialization` sections from context files - Automatic execution of startup actions like `recall_memory` and `search_by_tag` for memory integration - Session status tracking with execution time monitoring (< 2 second performance requirement) - Comprehensive error handling and logging for initialization actions - **Memory Service Integration Framework**: Foundation for mcp-memory-service integration - Built-in simulation layer for memory service calls (`recall_memory`, `search_by_tag`) - Memory retrieval results storage and analysis capabilities - Designed for easy replacement with actual mcp-memory-service calls - Smart context-aware memory integration patterns - **New MCP Tools**: Extended server with session management capabilities - `execute_session_initialization`: Manually trigger session startup actions - `get_session_status`: Retrieve detailed session initialization status and results - **Enhanced Context System**: - `memory_context.json`: Comprehensive memory service integration patterns - `date_awareness_context.json`: Temporal context awareness features - Session initialization configuration in existing context files ### Changed - **ContextProvider Architecture**: Extended with session state management - Added `session_status` tracking for initialization state - Enhanced singleton pattern with session-aware capabilities - Improved context loading with session initialization detection ### Technical Implementation - **Performance Optimized**: Session initialization completes in under 0.01 seconds - **Memory Integration Ready**: Framework supports real mcp-memory-service integration - **Error Resilient**: Graceful handling of failed initialization actions - **Extensible Design**: Easy addition of new session initialization actions ### Documentation - **CLAUDE.md**: Comprehensive development guide for Claude Code instances - **Test Framework**: Added `test_session_init.py` for development testing - **Architecture Documentation**: Updated with session initialization patterns ### Foundation for Future Features This release establishes the foundation for: - **Phase 2**: Dynamic context file creation and management via LLM tools - **Phase 3**: Synergistic integration with advanced learning capabilities - **Enterprise Features**: Team knowledge propagation and context analytics ## [1.3.0] - 2025-09-17 ### Added - **Dynamic Context Loading**: Implemented automatic discovery and loading of context files - Server now uses glob patterns to discover `*_context.json` files automatically - No more hardcoded context file lists - any new context files are automatically detected - Added support for `AUTO_LOAD_CONTEXTS` environment variable (defaults to `true`) - AppleScript context and any future contexts load automatically without code changes ### Changed - **Context Loading System**: Replaced hardcoded context file list with dynamic file discovery - `load_all_contexts()` method now scans the contexts directory using glob patterns - Improved error handling and logging for context file loading - Better debugging output showing discovered and loaded contexts ### Fixed - **AppleScript Context Loading**: AppleScript context now loads automatically as intended - Previously required manual addition to hardcoded list - Now discovered and loaded dynamically like all other contexts ### Technical Details - The server now properly implements the documented behavior of "automatically detects and loads any `*_context.json` files" - Environment variable `AUTO_LOAD_CONTEXTS=false` can disable auto-loading if needed - Improved logging shows context discovery and loading progress - Future-proof: new context files require no server code changes ## [1.2.1] - 2025-09-17 ### Fixed - **macOS Configuration Path Issue**: Fixed incorrect Claude Desktop configuration path detection - `install.sh`: Added proper macOS detection using `$OSTYPE` to use `~/Library/Application Support/Claude/` instead of `~/.config/claude/` - `verify_install.py`: Updated `find_claude_config()` function to use correct macOS path - Updated error messages to show platform-specific paths for better troubleshooting ### Changed - **Documentation Updates**: Corrected macOS paths throughout documentation - `README.md`: Separated Linux/macOS configuration file locations - `QUICKSTART.md`: Updated configuration paths and uninstall instructions - `TROUBLESHOOTING.md`: Fixed all references to use platform-specific paths for config files and logs - **Version References**: Updated all package references from 1.1.0 to 1.2.1 ### Technical Details - The installation scripts now correctly detect macOS and place configuration files in the proper location that Claude Desktop actually uses - Windows and Linux path detection were already correct and remain unchanged - This resolves the issue where the installer would create config files in the wrong location on macOS systems ## [1.2.0] - 2025-08-21 ### Fixed - **Issue #1**: DXT Package Installation Not Working as Documented - Removed references to non-existent `dxt install` command - Fixed bundled dependencies compatibility issues with Python 3.13+ - Added proper virtual environment-based installation approach ### Added - **Automated Installation Scripts**: Cross-platform installation automation - `install.sh`: Unix/Linux/macOS automated installer with error handling - `install.bat`: Windows automated installer with proper path handling - **Installation Verification**: `verify_install.py` script for comprehensive installation checks - **Quick Start Guide**: `QUICKSTART.md` with step-by-step installation instructions - **Enhanced Documentation**: - Updated README.md with accurate installation methods - Added platform-specific installation instructions - Improved troubleshooting documentation ### Changed - **Installation Process**: Now uses virtual environment approach instead of bundled dependencies - **Documentation Structure**: Reorganized installation options with automated scripts as primary method - **DXT Package Usage**: Updated to use `dxt unpack` instead of non-existent `dxt install` ### Technical Improvements - **Error Handling**: Comprehensive error detection and reporting in installation scripts - **Cross-Platform Support**: Proper path handling for Windows, macOS, and Linux - **Installation Verification**: Automated checks for Python version, dependencies, and configuration - **User Experience**: Clear progress indicators and helpful error messages ## [1.1.0] - 2025-01-08 ### Added - **DXT Package Distribution**: Complete Desktop Extension (DXT) installer for easy installation - Self-contained package with all Python dependencies bundled - Cross-platform support (Windows, macOS, Linux) - Automatic Claude Desktop configuration - Single-command installation: `dxt install mcp-context-provider-1.1.0.dxt` - **Enhanced Documentation**: Updated README.md with DXT installation instructions - **DXT Build System**: Added complete DXT directory structure and build process - **Distribution Ready**: 18.6 MB compressed package (44.6 MB unpacked) ### Changed - **Installation Methods**: Now supports both DXT (recommended) and manual installation - **README Structure**: Reorganized with DXT installation as primary method - **Package Metadata**: Updated manifest.json with comprehensive DXT schema compliance ### Technical Details - Package size: 18.6 MB compressed, 44.6 MB unpacked - Bundled dependencies: MCP SDK 1.13.0, Pydantic 2.11.7, and all required packages - Platform compatibility: Python 3.8+ on Windows, macOS, Linux - Installation time: < 30 seconds with DXT CLI ## [1.0.0] - 2025-01-08 ### Added - **Initial Release**: Complete MCP Context Provider implementation - **Core Server**: Python MCP server with STDIO JSON-RPC protocol - **Context System**: Tool-specific context rules and preferences - **Auto-Corrections**: Regex-based syntax transformations - **Persistent Context**: Survives Claude Desktop restarts and chat sessions ### Context Files - **DokuWiki Context**: Markdown to DokuWiki syntax conversion - **Terraform Context**: Resource naming conventions and best practices - **Azure Context**: Resource naming and compliance rules - **Git Context**: Commit message conventions and workflow patterns - **General Preferences**: Cross-tool preferences and standards ### MCP Tools - `list_available_contexts`: List all loaded context configurations - `get_context_rules`: Retrieve context rules for specific tool category - `apply_auto_corrections`: Apply automatic syntax corrections to text - `get_tool_preferences`: Get preferences for specific tool category ### Documentation - **README.md**: Comprehensive installation and usage guide - **CONTEXT_GUIDE.md**: Complete context file reference - **DEVELOPER_GUIDE.md**: Guide for creating custom contexts - **EXAMPLES.md**: Real-world usage examples - **TROUBLESHOOTING.md**: Common issues and solutions ### Technical Implementation - **MCP Protocol**: Full compliance with MCP 1.9.4+ specification - **Error Handling**: Robust error handling and validation - **Extensibility**: Easy addition of new context files - **Performance**: Fast startup and minimal resource usage - **Security**: No hardcoded secrets or credentials ### Testing & Quality - **Server Testing**: Comprehensive MCP server functionality tests - **Context Validation**: All context files validated against schema - **Integration Testing**: Verified with Claude Desktop integration - **Documentation**: Complete documentation suite with examples --- ## Version History Summary - **v1.2.0**: Fixed DXT installation issues and added automated installation scripts - **v1.1.0**: Added DXT package distribution and enhanced installation options - **v1.0.0**: Initial release with core MCP server and context system ## Upgrade Notes ### From 1.1.0 to 1.2.0 - **Breaking Change**: DXT bundled dependencies no longer recommended due to Python 3.13+ compatibility - **Migration**: Use new automated installation scripts or manual virtual environment setup - **Benefits**: More reliable installation across different Python versions and Linux distributions - **No Context Changes**: All existing context files and server functionality remain identical ### From 1.0.0 to 1.1.0 - **Existing Users**: Can continue using manual installation method - **New Users**: Recommended to use DXT installation for easier setup - **No Breaking Changes**: All existing context files and configurations remain compatible - **Enhanced Distribution**: DXT package provides better deployment and management options