mcp-taskflow
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DATA_DIR | No | Path to the data directory for persisting task state |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| plan_taskC | Plan tasks and construct a structured task list. Converts natural language descriptions into actionable task proposals with goals and expected outcomes. |
| analyze_taskC | Deeply analyze task requirements and propose a high-level approach. Includes threat modeling, design decisions, and architectural considerations. |
| reflect_taskC | Critically review analysis results and propose optimizations. Identifies potential issues, alternative approaches, and improvements. |
| split_tasksC | Split a complex task into a structured set of smaller tasks. Supports bulk creation with dependency graphs and various update modes. |
| list_tasksC | List tasks by status. Returns task overview with counts and filtering. |
| get_task_detailB | Get complete details for a specific task including dependencies and related files. |
| query_taskC | Search tasks by keyword or ID with pagination support. |
| update_taskC | Update task details, dependencies, or related files. |
| delete_taskB | Delete a task. Completed tasks are archived, others are removed. |
| clear_all_tasksB | Clear all tasks after confirmation. Creates backup before deletion. |
| execute_taskB | Mark a task as in progress and build an execution prompt with dependencies and implementation guide. |
| verify_taskC | Verify task completion with a score and summary. Marks task as completed. |
| init_project_rulesB | Initialize or update project coding rules and guidelines. |
| get_server_infoB | Get server metadata, version, and health status. |
| process_thoughtC | Record a structured thought step in a reasoning process with stage tracking. |
| research_modeC | Enter research mode to explore a programming topic in depth with iterative state tracking. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 16 tools
The CRUD-oriented task tools (delete_task, clear_all_tasks, list_tasks, get_task_detail, query_task, update_task, execute_task, verify_task) are reasonably distinct, but the cognitive cluster—reflect_task, analyze_task, plan_task, process_thought, research_mode—overlaps heavily and an agent could easily pick the wrong reasoning tool. list_tasks vs query_task also blur slightly.
Strong verb_noun snake_case pattern throughout (reflect_task, plan_task, delete_task, split_tasks, update_task, execute_task, verify_task). Minor deviations: research_mode is noun-style rather than verb-led, and get_task_detail adds a third token, but overall consistency is high.
16 tools is just above the comfortable range but each covers a plausible distinct operation within a task-management workflow. Slightly heavy, with some reasoning tools that could arguably be consolidated.
Good lifecycle coverage: task creation (plan_task/split_tasks), read (list_tasks, get_task_detail, query_task), update, delete, execute, and verify are all present, plus project rules and server info. Minor gaps around explicit single-task creation semantics and no dedicated dependency-graph query beyond get_task_detail.