Task Types

Open Source — GitScrum MCP Server is open source under the MIT license. Available on npm and GitHub. Model Context Protocol server for GitScrum — Claude, GitHub Copilot, Cursor, and any MCP-compatible client full operational access to your project management stack.

The task_type tool provides 4 actions for managing task types — the structural classification system that defines what kind of work each task represents. Task types like Bug, Feature, Improvement, and Technical Debt give your team a consistent way to categorize work items, enabling accurate reporting, sprint analysis, and workload distribution insights.

Task types in GitScrum are project-scoped, meaning each project maintains its own set of available types. This allows different projects to have different type taxonomies — a backend API project might use "Endpoint", "Migration", and "Bug", while a design project uses "Mockup", "Prototype", and "Review". The MCP Server lets your AI assistant discover available types, create new ones, and assign them when creating or updating tasks.

Actions Overview

Listing Task Types

The list action returns all task types configured for a specific project. Each type in the response includes its id, title, and color. The id field is what you pass as type_id when creating or updating tasks through the task tool.

This is typically the first call your AI assistant makes when it needs to create a task with a specific type — it fetches the available types, identifies the correct id, and uses it in the task creation call.

Required Parameters

Example Prompts

You:  "List available task types in the Backend project"
AI:   Calls task_type action=list → returns all types with IDs, titles, and colors

You:  "What task types can I use?"
AI:   Calls task_type action=list → presents the available type options

You:  "Show me the type configuration for the Frontend project"
AI:   Calls task_type action=list → displays the project's type taxonomy

Getting Task Type Details

The get action returns the complete details of a specific task type, including its title, color, and any additional metadata. This is useful when you need to inspect a type's configuration or verify its properties before making updates.

Required Parameters

Example Prompts

You:  "Show me the details of the Bug task type"
AI:   Finds the type via list → calls task_type action=get → returns full details

You:  "What color is the Feature type?"
AI:   Calls task_type action=get → extracts and presents the color value

Creating Task Types

The create action builds a new task type for the project. Task types need a title and can optionally include a color for visual identification. Once created, the type becomes available for assignment to any task in the project.

Required Parameters

Optional Parameters

Recommended Type Taxonomy

While every team can define their own types, these are common patterns that align with agile best practices:

Example Prompts

You:  "Create a new task type 'Technical Debt' with purple color"
AI:   Calls task_type action=create with title="Technical Debt", color="8B5CF6"

You:  "Add a 'Research' task type for the Backend project"
AI:   Calls task_type action=create with title="Research", company_slug, project_slug

You:  "Set up task types for a new project: Bug (red), Feature (blue),
       Improvement (green), and Chore (gray)"
AI:   Calls task_type action=create four times with distinct titles and colors

You:  "Create a 'Documentation' type in cyan"
AI:   Calls task_type action=create with title="Documentation", color="06B6D4"

Updating Task Types

The update action modifies an existing task type's title or color. Only the fields you specify are changed. This is useful for renaming types as your team's vocabulary evolves or adjusting colors for visual consistency.

Required Parameters

Optional Parameters

Example Prompts

You:  "Rename the 'Bug' type to 'Defect'"
AI:   Finds the type via list → calls task_type action=update with title="Defect"

You:  "Change the Feature type color to dark blue"
AI:   Calls task_type action=update with color="1D4ED8"

You:  "Update the Chore type to 'Maintenance' with a slate color"
AI:   Calls task_type action=update with title="Maintenance", color="64748B"

Using Task Types When Creating Tasks

Task types integrate directly with the task tool. When creating a task, you can pass the type_id parameter to assign a type. The AI assistant handles the lookup automatically — you mention the type by name, and the assistant resolves it to the correct numeric ID.

How Type Resolution Works

1. You say: "Create a bug task 'Login fails on Safari'"
2. The AI assistant calls task_type action=list to fetch available types
3. It finds the "Bug" type and extracts its id
4. It calls task action=create with type_id set to the Bug type's ID

This resolution happens behind the scenes. In practice, you simply mention the type by name in your prompt, and the MCP server handles the rest.

Example Prompts

You:  "Create a bug 'Memory leak in dashboard widget'"
AI:   Lists types → finds Bug ID → creates task with type_id

You:  "Add a feature task 'Dark mode support' to the Frontend project"
AI:   Lists types → finds Feature ID → creates task with type_id

You:  "Create a tech debt task to refactor the authentication module"
AI:   Lists types → finds Technical Debt ID → creates task with type_id

You:  "Log a research task 'Evaluate WebSocket libraries' with 4-hour estimate"
AI:   Lists types → finds Research ID → creates task with type_id and estimated_minutes=240

Task Type Reporting

Task types power the type distribution reports in sprint analytics. When you request sprint reports through the sprint tool's reports action with resource="types", the response breaks down tasks by their assigned type. This reveals how your team's effort is distributed across different kinds of work.

You:  "What percentage of our sprint is bug fixes vs features?"
AI:   Calls sprint action=reports with resource="types"
      → returns type distribution with percentages

You:  "Show me the type breakdown for the current sprint"
AI:   Calls sprint action=reports with resource="task_type_distribution"
      → returns detailed type distribution with completion status

This data is invaluable for retrospectives — if 60% of sprint work was bug fixes, the team might need to invest more in testing or code review to reduce defect rates in future sprints.

Context Auto-Resolution

The task type tool supports automatic context resolution. If the AI assistant already knows your workspace and project from a previous call in the conversation, it carries that context forward. This means you don't need to repeat project identifiers for every type operation.

Next Steps

• Tasks: Create tasks with specific types using the type_id parameter.
• Labels: Combine task types with labels for multi-dimensional categorization.
• Sprints: Analyze type distribution across sprints.
• Workflows: Configure the Kanban columns that tasks move through.
• Quick Start: Set up the MCP server if you haven't already.