MCP Claude Code

# Changelog All notable changes to the MCP Claude Code 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). ## [0.5.1] - 2025-08-29 ### Fixed - **Test Suite Compatibility** - Updated hidden files tests to reflect new .git directory access behavior - Fixed test expectations in test_various_hidden_files and test_real_world_project_paths - Tests now correctly expect .git paths to be allowed instead of excluded - Replaced .git exclusion test cases with node_modules exclusion test cases ## [0.5.0] - 2025-08-29 ### Added - **Generic Allow Pattern Override** (72ecafe) - New `--allow-pattern` CLI option to override default file/directory exclusions - Can be specified multiple times to allow access to typically excluded directories like node_modules, .git, etc. - Enhanced PermissionManager with `remove_exclusion_pattern()` method - Consolidated all default exclusions in PermissionManager for better organization - Updated DirectoryTreeTool to use PermissionManager exclusions consistently - Enhanced include_filtered functionality to properly traverse filtered directories ### Changed - **Permission System Updates** (fadf3e9) - Removed .git directory from default sensitive exclusions in PermissionManager - Tools now have visibility of .git contents by default unless explicitly restricted - Enhanced version control metadata access while maintaining other sensitive directory protections - Updated DirectoryTreeTool to reflect new .git access behavior ### Fixed - **Test Suite Updates** (4a02159) - Updated hidden files test to reflect .git/.vscode exclusions correctly - Fixed CI test failures in test_various_hidden_files - Moved .git/config and .vscode/settings.json to excluded_paths in tests - Updated test expectations to match correct exclusion behavior ## [0.4.0] - 2025-01-07 ### Added - **Persistent Shell Session Support** (32769b5) - New BashSession class for managing persistent shell environments with tmux - SessionManager and SessionStorage for comprehensive session lifecycle management - Session-based execution mode with environment persistence and working directory tracking - Support for interactive commands with `is_input` and `blocking` parameters - Command history tracking and session isolation - **Resource Management and Cleanup** (4c4a59d) - Background cleanup thread for periodic session expiration checks every 2 minutes - Signal handlers for graceful shutdown (SIGTERM, SIGINT) - Atexit handler for cleanup during normal termination - Reduced default session TTL from 30 to 5 minutes for aggressive cleanup - Session storage metrics and logging for cleanup operations - **Enhanced Cross-Platform Support** (184c2f7) - New CommandExecutor class for Windows-compatible shell command execution - Automatic platform detection with tmux availability checking - Support for multiple shell types (cmd, powershell, wsl) on Windows - Platform-specific command formatting and execution logic - **Project System Prompt Enhancements** (363f06f) - Factory function `create_project_system_prompt` for generating project-specific prompts - Manual project system prompt registration for on-demand access - Improved code organization and flexibility in prompt generation ### Changed - **BREAKING CHANGE: Shell Execution Architecture** (bf3233a) - Replaced CommandExecutor with BashSessionExecutor for all shell operations - All shell tools now use persistent sessions by default - Updated run_command tool interface with new session parameters - CommandResult now includes additional execution metadata - **BREAKING CHANGE: Tool Parameter Handling** (bc52019) - Migrated all tool parameters to TypedDict and Unpack for improved type safety - Consolidated parameter type definitions at module level - Enhanced IDE support and runtime type checking - Removed redundant parameter validation in favor of Pydantic handling - **Session ID Handling Improvements** (4263df1, 154a817) - Changed default session_id from None to empty string for consistency - Simplified session_id validation logic - Removed PROMPT_TEST echo command from session initialization - More predictable session handling behavior ### Removed - **Metadata System Simplification** (2cae74e) - Removed complex PS1 metadata system for command tracking - Eliminated CmdOutputMetadata class and PS1 prompt generation/parsing - Simplified CommandResult class to focus on core output - Improved reliability across different shell environments - **DocumentContext Removal** (db23971) - Removed DocumentContext class and all related functionality - Simplified codebase by removing redundant path tracking - Permission management now handled solely by PermissionManager - **Legacy Command Executor** (bf3233a) - Removed old CommandExecutor in favor of unified BashSessionExecutor - Consolidated shell execution logic into session-based approach ### Enhanced - **Command Output Isolation** (46c6c42) - Prevented command output accumulation in persistent sessions - Added automatic screen clearing and history management - Ensured clean execution context for sequential commands - **Platform-Specific Tool Registration** (184c2f7) - Automatic selection of appropriate run_command implementation - Seamless fallback between tmux-based and Windows-compatible execution - Maintained consistent tool interface across platforms ### Refactoring - **Improved Shell Execution Mode** (f862a6f) - Added subprocess execution mode to BashSessionExecutor - Enhanced flexibility in execution strategy selection - **Code Quality and Type Safety** (bc52019) - Comprehensive migration to modern Python type annotations - Improved developer experience with better IDE support - Reduced boilerplate validation code ## [0.3.4] - 2025-05-29 ### Added - New release script and process improvements (49c7deb) ### Refactored - **Tool Handler Architecture** (b1f5bda) - Simplified tool handler methods by using `get_context()` dependency injection - Removed explicit MCPContext parameter from all tool handler signatures - Improved code organization and reduced boilerplate across all tool implementations - Maintained functionality while making handler signatures cleaner and more consistent - **FastMCP Migration** (90a1419) - Updated all imports from `mcp.server.fastmcp` to `fastmcp` for improved consistency - Refactored tool parameter definitions to use pydantic Field annotations - Replaced deprecated parameter schema definitions with typed parameters - Enhanced type safety with TypedDict for complex parameter types - Added validation constraints through Field annotations - **Test Suite Cleanup** (801efbf) - Removed obsolete parameter schema validation tests following pydantic migration - Cleaned up deprecated mcp_description method tests - Simplified test suites to focus on core functionality rather than schema validation - Retained essential tool property assertions for name and description validation ## [0.3.2] - 2025-05-26 ### Fixed - **Project System Prompt Closure Bug** - Fixed critical closure bug in project system prompt registration where all prompts incorrectly used the last project path instead of their respective project paths - Multiple projects configured in MCP settings now correctly generate unique system prompts for each project - Replaced problematic closure with factory function pattern to ensure proper variable capture - Resolves issue where system prompts for all projects would reference 'claude-code-provider-proxy' regardless of the actual project being used ### Added - **Comprehensive Test Suite for Project System Prompts** - Added `tests/test_prompts.py` with 4 comprehensive test cases to prevent regression - Tests cover multiple project registration, single project registration, no projects scenario, and project basename handling - Includes specific test case that detects the closure bug to prevent future regressions - Mock-based testing approach ensures reliable verification of prompt generation functionality ## [0.3.1] - 2025-05-25 ### Fixed - **Todo ID Validation Compatibility** (da8295d) - Fixed todo ID validation to accept numeric IDs (integers and floats) in addition to strings - Resolves JSON parsing compatibility issue where numeric strings are automatically converted to integers - All IDs are now normalized to strings for consistent internal storage - Improved validation error messages for better debugging experience ### Maintenance - **Code Quality Improvements** (626ebec) - Removed unused imports and functions across multiple modules - Simplified validation.py by removing deprecated functions - Fixed linting issues including f-string formatting - Updated dependencies in uv.lock file - **Documentation Updates** (fc688ec) - Updated example documentation with latest functionality demonstrations - Added cross-references to tutorial documentation for version 0.3 features ## [0.3.0] - 2025-05-25 ### Added - **TODO Management System** (3f146ed) - Added TodoReadTool and TodoWriteTool for session-based task tracking - Implemented TodoStorage class for in-memory session-based storage - Support for todo items with status (pending/in_progress/completed) and priority (high/medium/low) - Session ID validation and isolation for secure multi-session support - Comprehensive test suite covering unit, integration, and validation scenarios - **Project Path Support and Prompt System** (2833d07) - Added new `--project` CLI argument to specify project paths for prompt generation - Refactored prompts into separate modules under `mcp_claude_code/prompts/` - Added PROJECT_SYSTEM_PROMPT for project-specific system prompt generation - Added utility functions for directory structure and git information extraction - Enhanced GitPython dependency for improved git integration - **MultiEdit Tool for Atomic Multi-File Operations** (2cd71bb) - Implemented atomic multi-edit operations with transaction semantics - Sequential edit application where each edit operates on previous results - Strict validation including exact string matching and expected replacement counts - File creation capability when first edit has empty old_string - Comprehensive diff generation and error handling - **Comprehensive Tutorial Documentation** (d71aea7) - Added new TUTORIAL.md with detailed instructions for advanced features - Project-specific system prompt application via Claude Desktop UI - Automated commands for context management and task continuity - Practical workflow scenarios and TODO management integration ### Enhanced - **TODO System Improvements** (f5c2766) - Added latest session tracking and auto-continue functionality - Renamed `continue_from_last_todo_list` to `continue_todo_by_session_id` for clarity - Added `continue_latest_todo` function for automatic session detection - Enhanced TodoStorage with timestamp tracking for session management - Added `find_latest_active_session` method for locating recent sessions - **Project TODO Reminder System** (ab3438d) - Implemented formatted todo list reminders for active sessions - Added empty state handling with PROJECT_TODO_EMPTY_REMINDER - Status indicators ([ ], [~], [✓]) and priority indicators (🔴, 🟡, 🟢) - Session isolation and comprehensive test coverage (569 lines) - **Tool Documentation and Usability** (Multiple commits) - Enhanced agent tool description with detailed usage guidelines (a989b39, 223c9d1) - Improved batch tool description with usage examples and scenarios (70266ff) - Enhanced grep_ast tool description with AST context examples (81b6658) - Updated thinking tool description with detailed use cases (c829dbf) - Clarified filesystem tool documentation for better user experience (ce6b09f, 9d5f144) ### Changed - **Tool Renaming and Consistency** (Multiple commits) - Renamed Jupyter notebook tools: 'read_notebook' → 'notebook_read', 'edit_notebook' → 'notebook_edit' (1be13ea) - Renamed 'edit_file' tool to 'edit' for consistency (8404eee) - Updated all references and documentation to reflect new naming conventions - **Installation and Documentation Improvements** (Multiple commits) - Completely rewrote installation guide with detailed instructions (56f28c3) - Updated INSTALL.md with improved project path configuration instructions (d71aea7, 8053705) - Added recommendation for setting projects individually for better system prompt generation - Simplified path parameter documentation and emphasized project path importance ### Removed - **Deprecated Tools and Features** (a1ee681) - **BREAKING CHANGE**: Removed `run_script` and `script_tool` functionality - Deleted implementation files and removed tool registration - Functionality consolidated into core `run_command` tool - **Deprecated File Operations** (d893f8b) - Removed GetFileInfoTool and all related references - Cleaned up batch tool and filesystem tool integrations - **Outdated Documentation** (Multiple commits) - Removed deprecated documentation files (USEFUL_PROMPTS.md, debug.md, migration guides) (8163f08) - Removed outdated system prompt documentation file (4eb866a) - Cleaned up redundant installation steps and outdated recommendations (42e7a69, 8be1a90) ### Refactoring - **Code Quality and Formatting** (49b4013) - Applied consistent code formatting and style improvements across codebase - Normalized whitespace, standardized quote usage, improved parameter alignment - Enhanced docstring formatting and removed unnecessary code complexity - **Configuration and Structure** (155992d) - Reorganized pyproject.toml configuration sections for better organization - Moved pytest.ini_options section and removed redundant configurations - **Tool Parameter Clarity** (6a14807) - Clarified session_id description to specify timestamp in seconds - Updated tool descriptions for better developer guidance ### Documentation - **Enhanced Installation and Configuration** (d71aea7, 56f28c3) - Multiple installation methods (uvx, pip/uv, development) - Detailed configuration options with practical examples - Agent tool setup instructions and performance optimization tips - Troubleshooting section and development procedures - **README Updates** (8ca5e04) - Added todo_write and todo_read commands to feature table - Updated tool references and removed outdated recommendations - **Cross-Reference Integration** - Added cross-references between installation and tutorial documents - Improved parameter table formatting and descriptions ## [0.2.2] - 2025-05-23 ### Added - Added command timeout configuration with `--command-timeout` CLI parameter (099401c) - Configurable timeout for shell command execution with default of 120 seconds - Increased default timeout from 30s to 120s for more reliable long-running operations - Enhanced shell tool reliability for complex command execution scenarios ### Enhanced - Improved error handling for network-related command execution (099401c) - Added `handle_connection_errors` decorator to gracefully handle client disconnections - Prevents MCP server crashes when running network-related commands that could cause disconnections - Enhanced connection stability for shell tools (run_command, run_script, script_tool) - Improved ToolContext logging to silently handle connection errors ### Changed - Optimized system prompt format and removed redundant instructions (f57e015) - Removed outdated sections from system prompt to improve efficiency - Leveraged Claude 4's enhanced capabilities to reduce unnecessary guidance - Streamlined prompt structure for better performance and clarity ### Fixed - Cleaned up code formatting by removing trailing whitespace (099401c) - Improved code quality and consistency across the codebase - Applied consistent formatting standards ## [0.2.1] - 2025-05-18 ### Fixed - Added proper single quote escaping for shell commands (2313d99) - Fixed command execution failures when commands contain single quotes - Modified `_execute_command` method to escape single quotes by replacing them with `'\''` - Added logging for original and escaped commands to assist with debugging - Added comprehensive tests for various quote patterns in shell commands - Tests verify proper handling of commands containing single quotes, double quotes, and mixed quote patterns ## [0.2.0] - 2025-05-16 ### Added - Added batch tool availability to dispatch_agent (c7bc6a8) - Enhanced agent capabilities by adding BatchTool to available tools - Enables batch processing for improved agent performance - Allows for concurrent execution of multiple operations ### Changed - Renamed read_files tool to read with improved functionality (42d8e94) - Simplified interface with single file_path parameter - Added line number display in output (cat -n format) - Added offset and limit parameters for reading specific line ranges - Added line truncation for very long lines - Improved error handling and parameter validation - Enforced absolute path requirement in agent prompts (c3fe452) - Added validation for absolute paths in dispatch_agent prompts - Improved error messages with usage examples - Enhanced reliability for filesystem operations - Renamed edit_file tool to edit throughout documentation and code (10cfe12) - Updated all references for consistency - Improved parameter handling and validation - Enhanced error messages ### Fixed - Removed outdated docstring section in agent_tool.py (c30ad80) - Eliminated obsolete information about args and returns - Maintained core tool functionality description - Updated directory verification tool reference in write.py docs (fb36e12) - Changed LS tool reference to directory_tree tool for accuracy - Aligned documentation with actual tool naming conventions ### Documentation - Added example for dispatching multiple agents in batch operations (8ed8769) - Demonstrated concurrent agent dispatch for information retrieval - Added batch operations as valid use case for agent operations - Clarified batch tool usage guidelines and added examples (9e14410) - Emphasized related changes vs. independent operations - Distinguished between simple mechanical changes and complex validations - Added practical usage scenarios and examples - Refactored dispatch_agent documentation (81292c8) - Simplified and consolidated usage guidelines - Updated concurrency recommendations for batch capabilities - Improved readability with clearer section organization ## [0.1.31] - 2025-05-16 ### Added - Added batch tool for parallel tool execution (eade6f7) - New tool that enables executing multiple tool invocations in parallel within a single request - Significantly improves performance by reducing context usage and latency for multiple operations - Supports independent tool operations with proper error handling and result formatting - Particularly useful for running multiple searches, file operations, or data gathering in parallel ### Changed - Renamed tool tags in system prompt for consistency and clarity (030956c) - Changed `<think_tool>` to `<think>` and `<grep_ast_tool>` to `<grep_ast>` to maintain consistent naming - Improves readability and aligns with the simpler tag naming pattern used elsewhere in documentation ## [0.1.30] - 2025-05-15 ### Added - Added release creation prompt and registration (98fbd45) - New prompt template with step-by-step guidance for creating project releases - Detailed instructions for version analysis, updates, changelog creation, and tagging - Structured approach for consistent release processes - Improved documentation for the compact() prompt function ## [0.1.29] - 2025-05-15 ### Added - Added compact conversation prompt and registration (893bef5) - New prompt module with detailed instructions for summarizing technical conversations - Structured format for capturing user requests, technical concepts, and problem-solving approaches - Helps maintain context during technical discussions with structured summaries ### Changed - Reorganized grep parameters and improved type hints (25c699a) - Reordered parameters to make tool_ctx required and first in parameter lists - Replaced Optional type hints with modern | None syntax - Added explicit type hints for path and include parameters - Improved code clarity while maintaining backward compatibility ### Maintenance - Added claude-code-doc directory to .gitignore (918e88e) - Prevents accidental commits of generated documentation files ## [0.1.28] - 2025-05-14 ### Added - Added liteLLM agent_base_url parameter for local LLM support (0f93398) - Enables integration with locally hosted models - Requires fake API key even when unnecessary - Enhanced CLI arguments for configuration ### Changed - Renamed SearchContentTool to Grep and integrated ripgrep for improved performance (132a603) - Added ripgrep integration for significantly faster file content searches - Enhanced pattern matching with regex support - Maintained backward compatibility through SearchContentTool shim - Renamed write_file tool to write and edit_file to edit for consistency - Added comprehensive documentation in migration_SearchContentTool_to_Grep.md ### Added - Added Windows shell execution and script support (d2a8441) - Support for CMD, PowerShell, and WSL shells - Enhanced command execution for cross-platform compatibility - Improved script handling for Windows environments ### Fixed - Added platform-specific support for temp folders (d1f60cf) - Improved cross-platform compatibility - Enhanced permissions system for Windows environments ## [0.1.27] - 2025-05-03 ### Added - Added AST-aware code search tool (`grep_ast`) that provides structural context for matches (6a37df6) - Shows how code matches fit within functions, classes, and other structures - Makes it easier to understand code organization when exploring unfamiliar codebases - Added dependency on grep-ast package - Includes comprehensive test suite covering various search scenarios ### Documentation - Updated system prompt with best practices for using the grep_ast tool - Enhanced README with grep_ast tool description - Added detailed usage instructions in tool documentation ## [0.1.26] - 2025-04-27 ### Fixed - Fix issue with not starting application due to litellm==1.67.2 (#11) ### Documentation - Correct formatting and clarify scope requirement in agent prompting guide (3532d4d) ## [0.1.24] - 2025-04-25 ### Changed - Simplified agent tool to support only single string prompt parameter - Renamed parameter from "prompts" to "prompt" for improved clarity - Removed multi-agent concurrent execution to simplify implementation - Updated documentation and tests to match new single-agent design ### Added - Added logging for agent tool execution results - Logs tool name and first 100 characters of result for improved debugging - Makes the execution flow more transparent in logs ### Documentation - Updated system prompt documentation for clarity - Clarified confirmation step by changing "before executing it" to "before executing any tool" - Enhanced thinking tool guidelines by emphasizing concise and accurate content - Refined learning step to specify documentation location in artifact - Removed redundant best practices that were covered by other principles - Removed deprecated /init command from user commands section - Updated Google model recommendation to gemini-2.5-flash-preview in documentation ### Refactoring - Removed --project-dir argument from CLI since project tools were removed ## [0.1.23] - 2025-04-19 ### Removed - Removed project tools package to streamline architecture - Removed `mcp_claude_code/tools/project` directory and related test files - Removed project analysis functionality which was redundant with filesystem tools - Updated import statements and tool registrations across codebase - Updated tests to maintain test coverage without project tools ### Changed - Simplified system prompt for improved usability - Removed redundant sections to improve Claude execution efficiency - Updated documentation to reflect current toolset - Replaced project analysis examples with code search examples ## [0.1.22] - 2025-04-19 ### Changed - Simplified permissions by removing unnecessary default exclusions from sensitive directories - Removed `.git`, `.config`, and `.vscode` from the default sensitive directories list - Improves usability while maintaining security for truly sensitive directories like `.ssh` and `.gnupg` ## [0.1.21] - 2025-04-19 ### Enhanced - Enhanced tool descriptions with parameter details in mcp_description method - Added support for displaying parameter descriptions from schema definitions - Improved formatting with hyphen separator for better readability - Updated tests to verify parameter descriptions are properly included - Makes tool descriptions more informative for Claude ### Fixed - Fixed pytest configuration by adding asyncio_default_fixture_loop_scope setting - Added explicit configuration in pyproject.toml to address deprecation warnings - Set asyncio mode to "strict" for more reliable async tests ### Documentation - Added project_info section to system prompt template - Includes repository name and owner details for better context - Improves system's ability to generate accurate project-specific responses ### Changed - Removed outdated TODO comment in agent tool registration - Updated .gitignore to exclude aider files and CLAUDE.md ## [0.1.20] - 2025-04-03 ### Added - Added new USEFUL_PROMPTS.md document with practical prompt templates for common scenarios: - Summarizing history for continued conversations - Automation research summaries - Release preparation commands - Resuming interrupted conversations ### Changed - Reduced default values for agent tool limits: - agent_max_iterations reduced from 30 to 10 - agent_max_tool_uses reduced from 100 to 30 - Enhanced system prompt documentation for the dispatch_agent tool: - Improved guidelines for when to use the dispatch_agent tool - Added detailed examples for effective agent prompting - Clarified agent limitations and capabilities - Added categorized use case recommendations ### Documentation - Improved README.md with clearer feature descriptions - Fixed formatting and alignment in README.md tables - Added cross-reference to USEFUL_PROMPTS.md in the README - Added specific guidance for file reading operations in system prompt ## [0.1.19] - 2025-03-28 ### Added - Added Agent Tool for task delegation and concurrent execution - Agent Tool enables Claude to delegate complex tasks to specialized sub-agents that can work in parallel, enhancing performance for complex operations like code search, analysis, and multi-step tasks - Agents use read-only tools to ensure safety while maintaining full access to information retrieval capabilities - Introduced new parameter `--enable-agent-tool` to allow users to control whether the agent tool is registered - Added parameters `--agent-max-iterations` (default: 30) and `--agent-max-tool-uses` (default: 100) to control agent behavior - Implemented multi-agent support to enable concurrent sub-agent execution for improved performance when handling multiple related tasks - Agents return results that are automatically formatted and integrated into Claude's responses - Added LiteLLM integration to support various model providers beyond OpenAI - Command line support for specifying the model name, API key, and token limits - Support for Anthropic, OpenAI, Google, and other LLM providers - Simplified configuration through environment variables or command-line arguments ### Enhanced - Enhanced system prompt with best practices and dispatch agent usage guidelines - Improved agent tool implementation with better parameter validation and error handling - Increased max_tool_uses limit from 15 to 30 for improved agent flexibility - Optimized agent implementation to avoid redundant checks and improve performance ### Changed - Restricted agent tool to use only read-only variants of filesystem tools for security - Simplified agent tool implementation to focus on multi-agent operation - Refactored system prompt handling and improved user prompt support - Updated parameter validation to support both string and array inputs for prompts ### Fixed - Fixed JSON serialization error in agent tool error handling - Fixed redundant type checks and improved test environment handling - Improved path description in read_files schema for better clarity - Updated type annotations to use built-in generics syntax ### Documentation - Added detailed documentation for the agent tool in INSTALL.md, explaining setup and configuration options - Added debugging guide for Model Context Protocol Inspector - Updated installation instructions with LLM model and token configuration options - Clarified cwd parameter restrictions in shell tool documentation ## [0.1.18] - 2025-03-24 ### Enhanced - Significantly improved performance of `search_content` and `content_replace` tools - Implemented parallel processing for file searching with batched execution - Optimized file finding strategy using more efficient directory traversal - Added semaphore-based concurrency control to avoid overwhelming the system ### Changed - Replaced recursive directory traversal with more efficient `pathlib.Path.rglob()` method - Restructured file search logic to filter allowed paths more efficiently - Changed file pattern matching to use `fnmatch` for more consistent behavior ## [0.1.17] - 2025-03-24 ### Enhanced - Enhanced `search_content` tool to use regular expression pattern matching instead of simple substring matching - Added comprehensive test suite for regex search functionality - Improved code search capabilities with pattern matching support ### Changed - Modified search implementation to use `re.search()` instead of the `in` operator - Maintained backward compatibility with existing string search patterns ## [0.1.16] - 2025-03-23 ### Added - Added Jupyter notebook support with `read_notebook` and `edit_notebook` tools - Implemented reading of notebook cells with their outputs (text, error messages, etc.) - Added capabilities for editing, inserting, and deleting cells in Jupyter notebooks - Added comprehensive test suite for the notebook operations ### Changed - Updated tools registration to include the new Jupyter notebook tools - Enhanced README.md with Jupyter notebook functionality documentation ## [0.1.15] - 2025-03-23 ### Added - Added support for file paths in `search_content` and `content_replace` tools - Extended functionality to search within and modify a single file directly - Implemented smarter path handling to detect file vs directory inputs ### Changed - Updated docstrings to clarify that `path` parameter now accepts both file and directory paths - Enhanced tool parameter descriptions for improved clarity ### Improved - Added extensive test coverage for the new file path functionality - Improved error messaging for file operations ## [0.1.14] - 2025-03-23 ### Added - Enhanced `directory_tree` tool with depth limits and filtering capabilities - Added parameter `depth` to control traversal depth (default: 3, 0 or -1 for unlimited) - Added parameter `include_filtered` to optionally include commonly filtered directories - Added statistics summary to directory tree output ### Changed - Improved directory tree output format from JSON to more readable indented text - Added filtering for common development directories (.git, node_modules, etc.) - Enhanced directory tree structure to show skipped directories with reason ## [0.1.13] - 2025-03-23 ### Changed - Improved README instructions regarding the placement of the system prompt in Claude Desktop - Clarified that the system prompt must be placed in the "Project instructions" section for optimal performance ## [0.1.12] - 2025-03-22 ### Changed - Modified permissions system to allow access to `.git` folders by default - Updated tests to reflect the new permission behavior ## [0.1.11] - 2025-03-22 ### Changed - Improved documentation in file_operations.py to clarify that the `path` parameter refers to an absolute path rather than a relative path - Enhanced developer experience by providing clearer API documentation for FileOperations class methods ## [0.1.10] - 2025-03-22 ### Added - Enhanced release workflow to reliably extract and display release notes ### Changed - Improved error handling and fallback mechanism for the release process ## [0.1.9] - 2025-03-22 ### Fixed - Fixed GitHub Actions workflow to properly display release notes from CHANGELOG.md - Added fallback mechanism when release notes aren't found in CHANGELOG.md ## [0.1.8] - 2025-03-22 ### Added - Added "think" tool based on Anthropic's research to enhance Claude's complex reasoning abilities - Updated documentation with guidance on when and how to use the think tool ## [0.1.7] - 2025-03-22 ### Fixed - Added validation in `edit_file` to ensure `oldText` parameter is not empty ## [0.1.6] - 2025-03-22 ### Fixed - Fixed GitHub Actions workflow permissions for creating releases ## [0.1.5] - 2025-03-22 ### Changed - Updated GitHub Actions to latest versions (v3 to v4 for artifacts, v3 to v4 for checkout, v4 to v5 for setup-python) ## [0.1.4] - 2025-03-22 ### Added - Added UVX support for zero-install usage ### Changed - Simplified README.md to focus only on configuration with uvx usage - Updated command arguments in documentation for improved clarity ## [0.1.3] - 2025-03-21 ### Fixed - Fixed package structure to include all subpackages - Updated build configuration to properly include all modules ## [0.1.2] - 2025-03-21 ### Added - Published to PyPI for easier installation - Improved package metadata ## [0.1.1] - 2025-03-21 ### Added - Initial public release - Complete MCP server implementation with Claude Code capabilities - Tools for code understanding, modification, and analysis - Security features for safe file operations - Comprehensive test suite - Documentation in README ### Changed - Improved error handling in file operations - Enhanced permission validation ### Fixed - Version synchronization between package files ## [0.1.0] - 2025-03-15 ### Added - Initial development version - Basic MCP server structure - Core tool implementations