task-manager-mcp
If you are the rightful owner of task-manager-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.
Task Manager MCP Server is designed to enhance AI code editors like Cursor by providing an unlimited context window and task management capabilities.
The Task Manager MCP Server is a robust solution for integrating with AI code editors, particularly Cursor. It leverages the capabilities of Gemini 2.5 and Cursor to provide a seamless task management experience with an unlimited context window. Inspired by Roo Code's Boomerang mode, this server offers a cost-effective alternative by utilizing Cursor's subscription model. The server includes a Svelte UI for task management, allowing users to view, adjust, and review tasks efficiently. It supports complex feature planning, conversation history tracking, and a clarification workflow to ensure smooth interaction between the user and the AI. Additionally, it offers task CRUD operations and an optional automatic code review feature.
Features
- Complex Feature Planning: Generates a step-by-step coding plan using LLM with project context, breaking down high-effort tasks recursively.
- Integrated UI Server: Runs an Express server to serve static frontend files and provides API endpoints for the UI, displaying task lists and progress.
- Unlimited Context Window: Utilizes Gemini 2.5's 1 million token context window with truncation when needed.
- Conversation History: Tracks conversation history for each feature in separate JSON files, providing context for adjustments.
- Clarification Workflow: Pauses planning when more information is needed, interacting with the UI via WebSockets.
Tools
plan_feature
Generates a plan for a given feature request.
mark_task_complete
Marks a task as complete and retrieves the next task.
get_next_task
Fetches the next task in the queue.
review_changes
Analyzes code changes and creates new tasks if necessary.
adjust_plan
Allows for adjustments to the task plan.