Skip to main content
Glama

TaskFlow MCP

CHANGELOG.md9.65 kB
# Changelog All notable changes to the TaskFlow MCP 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). ## [1.5.0] - 2025-09-15 ### BREAKING CHANGES - **Task Interface Simplified** - Removed unused `approved` field from Task interface (Issue #12) - The `approved` boolean field was never used by any MCP tools and was blocking archiving functionality - Tasks are now considered complete when `done: true` without requiring separate approval - This aligns with the simplified workflow introduced in v1.3.2 - **Migration**: Any existing task files with `approved` fields will continue to work, but the field will be ignored - **Per-Request Task Numbering** - Changed from global to per-request task numbering (Issue #13) - Tasks now use per-request numbering: req-1 has task-1, task-2; req-2 has task-1, task-2, etc. - Subtasks also use per-request numbering: subtask-1, subtask-2 per request - More intuitive and aligns with user expectations of independent request workflows - **Migration**: Existing task files will continue to work, but new tasks will use the new numbering scheme ### Fixed - **Archiving Workflow** - Archive operations no longer blocked by unused approval requirements - **Task ID Clarity** - Task numbering now matches user mental model of request-scoped workflows ### Improved - Better separation between global counters (for notes) and per-request counters (for tasks) - Cleaner task interface without unnecessary fields - More intuitive task management workflow ## [1.4.2] - 2025-09-08 ### Added - **Request Completion Tool** - New `mark_request_complete` MCP tool to mark entire requests as completed (Issue #10) - Validates that all tasks and subtasks are completed before allowing request completion - Required before requests can be archived - Supports optional completion details/notes - Automatic archiving integration when `ARCHIVE_MODE=auto-on-complete` - Comprehensive error handling for incomplete tasks or subtasks ### Fixed - **Request Completion Workflow** - Addressed issue where requests couldn't be marked as complete (Issue #10) - Users can now properly complete requests which enables proper archiving workflow - Request completion status is now properly tracked and validated - Archive operations now work correctly with completed requests - **Archive File Auto-Creation** - Confirmed and documented automatic archive file creation (Issue #11) - Archive files are automatically created when archiving is performed - No manual file creation required by users - Proper directory structure creation with error handling - Existing functionality was working correctly, issue was resolved through clarification ### Improved - Enhanced documentation with new `mark_request_complete` tool usage examples - Updated README.md with complete workflow documentation - Better error messages for request completion validation ## [1.4.1] - 2025-09-03 ### Added - **Archive System** - Complete task archiving functionality (Issue #9) - `archive_completed_requests` tool to archive finished requests and keep active tasks clean - `list_archived_requests` tool to browse archived requests with search and filtering capabilities - `restore_archived_request` tool to restore archived requests back to active status - Archive files include completion timestamps, metadata, and full request history - Comprehensive error handling and validation for archive operations - Archive file structure with version tracking and metadata management - **Prompts System** - Global prompts configuration management (Issue #8) - `get_prompts` tool to retrieve current prompts configuration - `set_prompts` tool to set global instructions, taskPrefix, and taskSuffix - `update_prompts` tool to update specific parts of prompts configuration - `remove_prompts` tool to remove entire configuration or specific fields - Prompts are stored alongside tasks in the same file (JSON/YAML) - Automatic timestamp tracking for prompts creation and updates - Backward compatibility with existing task files without prompts - **Enhanced Path Resolution** - Improved file path handling (Issue #7) - Support for relative paths in TASK_MANAGER_FILE_PATH environment variable - Automatic path resolution using Node.js path.resolve() and path.normalize() - Cross-platform compatibility for Windows and Linux path formats - Support for TASK_MANAGER_BASE_DIR to override default base directory - Comprehensive error handling and fallback mechanisms for path resolution - Enhanced documentation with path usage examples and best practices ### Improved - Enhanced progress tracking with better task status visualization - Improved error messages and validation across all tools - Better handling of edge cases in task and subtask management - More robust file operations with proper error recovery ### Fixed - Path resolution issues on different operating systems - Task completion status validation in archive operations - Memory management improvements in large task sets ## [1.3.3] - 2025-01-05 ### Added - YAML format support for task data persistence - Tasks can now be saved in YAML format by configuring the file path with `.yaml` or `.yml` extension - Automatic format detection based on file extension during load and save operations - YAML format better preserves multiline descriptions and complex text content - Maintains backward compatibility with existing JSON format ### Fixed - Robust newline handling in task descriptions and subtask descriptions - Fixed corrupted newline patterns like 'nn' and 'n-' that were appearing in saved data - Implemented comprehensive string sanitization to normalize various newline formats - Improved sanitization logic to be more precise and avoid false positives with legitimate text - Fixed specific issue #4 where "instructions.mdnnFiles" and "analyze:n-" patterns were corrupted - Ensured proper preservation of multiline content in both JSON and YAML formats - Fixed issues where newlines were being incorrectly escaped or replaced during save operations ## [1.3.2] - 2025-08-19 ### Changed - Simplified approval workflow to use direct user confirmation instead of dedicated approval tools - Updated tool descriptions to clarify the approval process - Streamlined progress reporting by removing approval status columns ### Removed - Removed unused approval tool schemas (`ApproveTaskCompletionSchema` and `ApproveRequestCompletionSchema`) - Removed approval status tracking from progress tables and status reports - Removed approval counts from request summaries ### Fixed - Fixed workflow clarity by removing references to non-existent approval tools - Improved user experience by eliminating confusion around approval tool calls - Enhanced documentation consistency across all tool descriptions ## [1.2.1] - 2025-05-10 ### Fixed - Updated documentation to reflect current features and functionality - Fixed minor issues in the README.md and OVERVIEW.md files - Improved example system prompts for better clarity ## [1.2.0] - 2025-05-09 ### Added - Export functionality - Added `outputPath` parameter to `plan_task` tool to save task plan as Markdown - New `export_task_status` tool to export current task status in Markdown, JSON, or HTML format - Detailed task plan export with checkboxes, dependencies, and notes - Status report export with progress bars and detailed task information - Dependencies management - Added support for project and task-level dependencies - Dependencies can include name, version, URL, and description - New `add_dependency` tool to add dependencies to requests or tasks - Dependencies are displayed in progress tables and exports - Notes management - Added support for project-level notes - Notes can include title, content, and timestamps - New tools for managing notes: - `add_note`: Add a note to a request - `update_note`: Update an existing note - `delete_note`: Delete a note from a request - Notes are displayed in progress tables and exports ## [1.1.0] - 2025-05-09 ### Added - Subtasks functionality - Tasks can now have subtasks, which are smaller units of work - All subtasks must be completed before a task can be marked as done - New tools for managing subtasks: - `add_subtasks`: Add subtasks to an existing task - `mark_subtask_done`: Mark a subtask as completed - `update_subtask`: Update a subtask's title or description - `delete_subtask`: Delete a subtask from a task - Updated progress table to display subtasks with their status - Enhanced task details to include subtask information ## [1.0.0] - 2025-05-09 ### Added - Initial release of TaskFlow MCP - Core TaskFlowServer class for managing tasks and requests - MCP Server integration using @modelcontextprotocol/sdk - Data persistence using JSON file storage - Tool definitions for AI assistants to interact with the task manager - Task planning functionality - Task progress tracking with visual tables - User approval workflow for tasks and requests - Ability to add, update, and delete tasks - Detailed reporting of task status and progress - Configuration options for task file path - MCP configuration for easy integration with AI assistants ### Changed - N/A (Initial release) ### Deprecated - N/A (Initial release) ### Removed - N/A (Initial release) ### Fixed - N/A (Initial release) ### Security - N/A (Initial release)

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/pinkpixel-dev/taskflow-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server