Sprints

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 sprint tool provides 10 actions that cover the complete sprint lifecycle — from creating and planning sprints to tracking progress, analyzing performance metrics, and reviewing detailed reports. Every sprint operation your team performs in the GitScrum web application is available through natural language conversation with your AI assistant.

Sprints in GitScrum are time-boxed iterations that organize tasks into focused delivery cycles. The MCP Server gives your AI assistant the ability to create sprints, monitor velocity, review burndown charts, and generate performance reports — all without leaving your IDE or AI client.

Actions Overview

Listing Sprints

The list action returns all sprints within a specific project, including their slugs, titles, date ranges, and status. Each sprint in the response includes its slug, which you need for all subsequent sprint operations.

The all action returns sprints across all your workspaces and projects. This is useful when you need a cross-project view of sprint timelines or want to find a sprint without knowing which project it belongs to.

You:  "List all sprints in the Backend project"
AI:   Calls sprint action=list → returns sprint list with slugs, dates, and status

You:  "Show me sprints across all my projects"
AI:   Calls sprint action=all → returns cross-workspace sprint overview

Creating Sprints

The create action builds a new sprint with title, dates, and optional configuration in a single call. If you omit dates, the sprint defaults to starting today with a 7-day duration.

Required Parameters

Optional Parameters

Example Prompts

You:  "Create a sprint 'v2.1 Release' for the next 2 weeks"
AI:   Calls sprint action=create with title="v2.1 Release",
      date_start=[today], date_finish=[today + 14 days]

You:  "Start a new 1-week sprint called 'Sprint 14' in the Backend project"
AI:   Calls sprint action=create with title, company_slug, project_slug,
      date_start, date_finish

You:  "Create a private sprint 'Security Audit' that auto-closes on Feb 28"
AI:   Calls sprint action=create with title, is_private=true,
      close_on_finish=true, date_finish="2026-02-28"

Updating Sprints

The update action modifies any sprint property. Only the fields you specify are changed — everything else remains untouched. This is useful for extending deadlines, updating descriptions, or adjusting sprint settings mid-cycle.

Required Parameters

All optional parameters from create are available in update.

You:  "Extend the current sprint by 3 days"
AI:   Gets current sprint → calls sprint action=update with new date_finish

You:  "Rename the sprint to 'Q1 Stabilization'"
AI:   Calls sprint action=update with title="Q1 Stabilization"

You:  "Add a description to the sprint: 'Focus on performance improvements'"
AI:   Calls sprint action=update with description

Sprint Details

The get action returns the complete sprint object, including all configuration, dates, task counts, and metadata. This is the starting point when you need detailed information about a specific sprint.

You:  "Show me the details of Sprint 14"
AI:   Calls sprint action=get → returns full sprint data including dates,
      task counts, and configuration

You:  "What's the current sprint about?"
AI:   Finds the current/active sprint → calls sprint action=get

Sprint Analytics

The sprint tool includes four analytics actions that provide different perspectives on sprint performance. Each returns structured data that the AI assistant can summarize, compare, or analyze.

KPIs

The kpis action returns key performance indicators for a sprint — metrics like completion rate, velocity, scope changes, and on-time delivery percentage. These are the numbers a project manager checks in the sprint dashboard.

You:  "What are the KPIs for the current sprint?"
AI:   Calls sprint action=kpis → returns completion rate, velocity, scope metrics

You:  "How is Sprint 14 performing compared to our goals?"
AI:   Calls sprint action=kpis → AI analyzes the metrics against targets

Stats

The stats action returns a statistical summary — total tasks, completed tasks, remaining work, distribution by status, and assignee breakdown. This is the data behind the sprint board's summary widgets.

You:  "What's the current sprint status?"
AI:   Calls sprint action=stats → returns task counts by status, assignee distribution

You:  "How many tasks are left in the sprint?"
AI:   Calls sprint action=stats → extracts remaining task count

Progress

The progress action returns real-time completion progress as a percentage, along with task movement data. This is the sprint progress bar you see in the GitScrum web application, exposed as structured data.

You:  "How far along is the current sprint?"
AI:   Calls sprint action=progress → returns percentage complete, tasks in each state

You:  "Are we on track to finish the sprint on time?"
AI:   Calls sprint action=progress → AI analyzes velocity vs remaining work

Metrics

The metrics action returns delivery metrics including velocity (story points or task count per sprint), throughput, cycle time, and lead time. These are the metrics used for capacity planning and retrospectives.

You:  "What's our team velocity this sprint?"
AI:   Calls sprint action=metrics → returns velocity data

You:  "Show me the cycle time for Sprint 14"
AI:   Calls sprint action=metrics → extracts cycle time metrics

Sprint Reports

The reports action generates detailed, chart-ready sprint reports. You can request specific report types or get all available reports at once.

Report Types

Parameters

You:  "Show me the burndown chart for the current sprint"
AI:   Calls sprint action=reports with resource="burndown" → returns daily data points

You:  "Generate a full sprint report for Sprint 14"
AI:   Calls sprint action=reports without resource → returns all report data

You:  "How is the workload distributed across the team?"
AI:   Calls sprint action=reports with resource="member_distribution"

You:  "Show me the types of tasks in this sprint"
AI:   Calls sprint action=reports with resource="types"

Sprint Planning Workflow

For teams using MCP-driven sprint management, this workflow covers the complete sprint lifecycle:

1. Create the sprint

Define the sprint boundaries with a title and date range.

You:  "Create a 2-week sprint called 'Sprint 15' starting Monday"
AI:   Creates the sprint with calculated start and end dates

2. Review the backlog

Check available tasks before assigning work to the sprint.

You:  "Show unassigned tasks in the Backend project"
AI:   Calls task action=filter with unassigned=true → shows available work

3. Add tasks to the sprint

Assign tasks to the newly created sprint.

You:  "Add the top 5 priority tasks to Sprint 15"
AI:   Filters by priority → updates each task with sprint_slug

4. Monitor progress

Track the sprint daily for blockers and progress.

You:  "How is Sprint 15 going?"
AI:   Calls sprint action=progress + action=stats → summarizes current state

You:  "Are there any blockers in the sprint?"
AI:   Calls task action=filter with is_blocker=true and sprint="Sprint 15"

5. Review reports

Analyze sprint performance for retrospectives and capacity planning.

You:  "Generate the Sprint 15 retrospective data"
AI:   Calls sprint action=reports → returns burndown, velocity, and distribution data

You:  "Compare our velocity across the last 3 sprints"
AI:   Calls sprint action=metrics for each sprint → AI compares velocity trends

Context Auto-Resolution

The sprint tool supports automatic context resolution. If the AI assistant already knows your workspace and project from a previous call in the conversation, you can omit companyslug and projectslug — the MCP server resolves them from the conversation context.

This means you can say "Create a sprint" without specifying the project every time, as long as the AI assistant established that context earlier in the conversation.

Cross-Workspace Sprints

The all action provides a unique cross-workspace view of sprints. This is particularly valuable for agency owners and managers who oversee multiple projects across different workspaces. The response includes workspace and project identifiers for each sprint, allowing the AI assistant to drill into any specific sprint from the aggregated view.

You:  "Show me all active sprints across my workspaces"
AI:   Calls sprint action=all → returns sprints from all projects and workspaces

You:  "Which sprints are ending this week?"
AI:   Calls sprint action=all → AI filters by date_finish in current week

Next Steps

• Tasks: Create and manage the tasks that live inside sprints.
• Projects: Configure project workflows, types, and team members.
• Time Tracking: Track time spent on sprint tasks.
• Quick Start: Set up the MCP server if you haven't already.