GitScrum Tasks MCP for AI. Manage your full backlog without switching apps.
Works with every AI agent you already use
…and any MCP-compatible client








Connect to your AI in seconds.
GitScrum Tasks connects your project management workflow directly to any AI agent. It lets you manage the entire task lifecycle—creating new items, adding subtasks and checklists, assigning ownership, setting story points, and tracking progress—all conversationally.
Stop leaving your IDE or chat interface to update a Jira board; this MCP handles complex team coordination right where you are working.
What your AI can do
Assign member
Adds or removes a user from being responsible for a task.
Create checklist item
Adds a new step to a checklist attached to an existing task.
List checklists
Shows all the checklist items currently associated with a given task.
You can query tasks using advanced filters for status, assignee, label, or date range.
The agent links existing tasks as subtasks and retrieves related items across the project.
You can add checklist items, mark them done or undone, and toggle a main task's completion status.
The agent adds rich comments to a task or assigns/unassigns team members instantly.
You can create new tasks, update existing ones, duplicate them, and move them between projects.
Ask an AI about this
Waiting for input…
GitScrum Tasks: 28 Available Tools
These tools let your agent handle every step of the task lifecycle, from creating initial records to updating completion status and tracking relationships.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using GitScrum Tasks on VinkiusAssign Member
Adds or removes a user from being responsible for a task.
Create Checklist Item
Adds a new step to a checklist attached to an existing task.
List Checklists
Shows all the checklist items currently associated with a given task.
Toggle Checklist Item
Marks a specific item on a checklist as completed or undone.
Create Comment
Posts a detailed comment on a specific task, supporting rich text formatting.
Delete Comment
Removes an existing comment from a task's history.
List Comments
Retrieves and displays the full history of comments on a specific task.
Update Comment
Edits the text of an existing comment on a task.
Duplicate Task
Creates an exact copy of an existing task for reuse elsewhere.
List Effort Levels
Shows all available effort or priority levels used in the project.
Set Task Estimate
Sets a story point value or effort estimate for any given task.
Move Task To Project
Changes the main project that a task belongs to.
My Tasks
Gets a list of all tasks that are currently assigned directly to you.
Related Tasks
Finds and lists tasks that have an established connection to the current task.
Link Subtask
Attaches an existing task to another task as a required subtask item.
List Subtasks
Displays all subtasks linked to a main task.
Unlink Subtask
Separates a subtask from its parent task.
Get Task By Code
Finds and retrieves a task using its human-readable code, like WEB-42.
Create Task
Creates and initializes an entirely new task within the project.
Delete Task
Permanently removes a task from the system. This action cannot be undone.
Toggle Task Done
Changes the main completion status of an entire task (open to done).
Get Task
Retrieves all details about a specific task using its unique identifier (UUID).
List Tasks
Fetches a list of tasks using advanced filters like status, sprint, or assignee.
Update Task
Modifies general details and metadata for an existing task.
My Today Tasks
Retrieves a list of tasks due for completion today.
Create Task Type
Defines a new category or type for tasks (e.g., Bug, Chore) with its own color code.
List Task Types
Lists every defined task type available within the project, along with their colors.
Unassign Member
Removes a user from the list of people assigned to a task.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with GitScrum Tasks, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by GitScrum. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 28 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
The status board always feels out of sync.
Today, updating a task means jumping between five different places. You click the task card in Jira to update its status; then you go back to Slack to ping your teammate for sign-off; then you open GitHub to add a comment with code snippets. It's manual, it involves copy/pasting data, and it always feels like something is missed.
With this MCP, you just talk to your agent. You tell it: 'Change WEB-42 status to done, notify John, and add the test results.' The entire multi-step process—updating status, commenting, assigning people—happens conversationally in one spot. What you get is a single source of truth that updates instantly.
GitScrum Tasks MCP: Full Task Lifecycle Control
You no longer need to manually create the task, then find its UUID, and *then* open a separate tool just to assign someone. The agent handles creating the 'create_task' record and simultaneously running 'assign_member' using one prompt.
It’s about making the workflow invisible. You focus on the work itself, not the administrative steps of tracking it down. This is how fast development teams operate now.
What your AI can actually do with this
You shouldn't have to switch apps just to move a task along the board. This connector lets your agent handle all of that manual clicking for you. You can ask it to create an entirely new bug or feature, assign it to a specific developer, and even set story points without opening a browser tab.
Need to review progress? Your agent pulls up tasks based on filters like 'in-progress' or 'high effort,' showing exactly who owns what. It handles the messy details too: adding checklist items for investigation steps, logging detailed comments, or duplicating an existing task into a new project. With Vinkius in your catalog, you get one connection point to manage every aspect of your sprint backlog conversationally.
You just talk to your agent, and it manages the whole flow.
019d8441-ecee-7282-8d26-fb9b0f0c8d54 Here's how it actually works
The bottom line is that you never have to leave your chat interface to update project details again.
Subscribe to this MCP through Vinkius and enter your GitScrum API token and company slug.
Give your agent a natural language request, like 'Create a new tech debt task for the payments module.'
Your agent processes the request using available tools, creating or modifying the task status and reporting back what was done.
Who is this actually for?
Developers who hate context switching, Scrum Masters tired of manual board updates, and Product Managers who need instant visibility into sprint bottlenecks.
Needs to quickly create a bug task or mark checklist items complete without leaving the IDE.
Manages sprint backlogs by assigning tasks, setting estimates, and moving items between workflow columns via chat commands.
Filters and reviews task progress across multiple dimensions—like filtering all 'high effort' tasks assigned to a specific team member.
What Changes When You Connect
You never have to context switch to the GitScrum UI again. Your agent handles every task update, assignment, and status change conversationally.
Review complex progress instantly. You can ask for a list of tasks filtered by multiple dimensions—status, effort level, or assignee—all in one query using 'list_tasks'.
Keep your history clean. Instead of manually updating comments, you just tell your agent to create or update the comment, maintaining rich context.
Manage dependencies effortlessly. If a task needs further work, use the agent to link an existing item as a subtask or retrieve all related tasks using 'related_tasks'.
Track effort and ownership precisely. Use 'set_task_estimate' to assign story points and 'assign_member' to ensure every piece of work has clear ownership.
Handle cleanup quickly. You can delete tasks permanently with 'delete_task', or remove users from a task using 'unassign_member'.
See it in action
The sprint review is running long and you need status updates.
A PM asks their agent: 'Show me all tasks in the 'In Progress' column that are high effort.' The agent calls 'list_tasks' and presents a filtered list, letting the PM immediately identify bottlenecks without opening the board.
You finished unit testing but need to log results and assign it.
A developer tells their agent: 'Mark WEB-42 as done, add a comment saying I passed tests, and assign it to QA.' The agent executes 'toggle_task_done', then 'create_comment', and finally 'assign_member'—all in one chat sequence.
A feature requires several investigation steps before coding can start.
The Scrum Master asks the agent to add a checklist: 'Add a checklist for WEB-51 with items like 'Review API docs' and 'Get stakeholder sign-off.' The agent uses 'create_checklist_item,' turning vague goals into measurable next steps.
You are duplicating a component across two different products.
Instead of manually recreating the entire item, the developer asks the agent to duplicate WEB-105 and move it to the new project. The agent calls 'duplicate_task' followed by 'move_task_to_project', saving time.
The honest tradeoffs
Over-relying on manual API calls
A user tries to update task status, then separately sends a message to create a comment. They forget that the two actions must be related.
Don't send separate commands. Ask your agent conversationally: 'Update WEB-42 to done and leave a note saying I finished it.' The agent handles both 'toggle_task_done' and 'create_comment' in sequence.
Forgetting task relationships
A user wants to track the investigation steps for WEB-42 but only remembers listing all tasks, which gives them too much noise.
Tell your agent you want a checklist. It calls 'list_checklists' and uses 'create_checklist_item' to add specific, actionable sub-steps relevant only to that task.
Using UUIDs when human codes are better
A user has to copy a 32-character UUID from a dashboard and paste it into the prompt, which is slow and error-prone.
Just tell your agent the task code. It calls 'get_task_by_code' so you can reference tasks using easy names like WEB-42.
When It Fits, When It Doesn't
Use this MCP if your team needs to manage complex, multi-stage project workflows where status updates and collaboration are constant. You should use it when the primary bottleneck is 'context switching'—meaning you spend more time clicking between Jira, Slack, and your IDE than actually coding or planning.
Don't use this if you only need simple task lists without dependencies (e.g., just viewing names). If all you want to do is retrieve basic metadata for a few tasks, a simpler data retrieval tool might suffice. But because this MCP handles everything from 'list_tasks' to setting estimates with 'set_task_estimate', it’s built for deep process integration.
Questions you might have
How do I list tasks using the GitScrum Tasks MCP? +
You ask your agent to 'list_tasks' and provide filters like status, sprint, or assignee. It will return a filtered view of all relevant items.
Can I add comments with the GitScrum Tasks MCP? +
Yes. You use the tool to create comments on any task, supporting rich text so you can include code blocks or formatting.
What is 'set_task_estimate' for in the GitScrum Tasks MCP? +
This tool allows you to assign a story point value or effort estimate to any task, helping the team predict how long it will take.
Can I find a task without its UUID using the GitScrum Tasks MCP? +
Absolutely. You use 'get_task_by_code' and just give your agent the human-readable code, like WEB-42, to fetch all details.
Does the GitScrum Tasks MCP handle subtasks? +
Yes. You can use 'link_subtask' to attach existing tasks as dependencies and list them with 'list_subtasks'.
How do I modify details on an existing task using the `update_task` tool? +
It lets you change almost any field on a task, like its description or due date. You must provide the unique task UUID and a payload detailing exactly what needs updating.
What happens when I use the `delete_task` tool? +
The task is permanently removed from GitScrum; this action cannot be undone. Always verify the task ID before executing this command to prevent accidental data loss.
How do I pull a list of tasks assigned only to me with `my_tasks`? +
This tool fetches every task currently assigned to your account immediately. It’s the fastest way to get an overview of your current workload without needing complex filtering.
Can my AI agent find a specific task by its human-readable code like WEB-42? +
Yes! Use the get_task_by_code tool with the task code (e.g., WEB-42). Your agent will return the full task details including description, assignees, workflow status, labels, sprint association, time estimates, and checklist progress — in seconds.
I want to quickly see what I need to work on today. Can the agent help? +
Absolutely. Use my_today_tasks to see everything due today, or my_tasks for your full assignment list across all workspaces. The agent returns task titles, project context, due dates, and current workflow status — your daily standup briefing in one command.
Can the agent manage task assignments and checklists too? +
Yes. Use assign_member and unassign_member to manage task assignments by username. For checklists, use create_checklist_item to add items (including nested sub-items via parent_id), and toggle_checklist_item to mark them done. The complete task lifecycle is covered — from creation to completion.
We've already built the connector for GitScrum Tasks. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 28 tools are live and waiting.
You're up and running in seconds.
Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.