Tower MCP. Manage projects and tasks via natural conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Tower MCP Server gives your AI agent direct access to a team's entire project lifecycle. It lets you manage tasks, track projects, and read discussions without ever touching the web UI.
Your agent can list all active projects, create new tickets, update status reports using `update_task`, and even pull up shared documents via `list_doc_folders`.
This is pure command-line productivity for project management.
What your AI agents can do
Create task
Generates a brand new task item within a specified project.
Get project
Retrieves detailed information about a single, specific project workspace.
Get task details
Pulls all the specifics—assignee, description, status—for one existing task item.
Run list_projects to get a real-time roster of every collaboration workspace you have access to.
Use create_task or update_task to instantly generate new work items or change the status/assignee on existing ones.
Run list_members to get a list of all available users and teams for assignment purposes.
Call list_discussions to pull up the latest topics and conversations happening within any given project.
Run list_doc_folders to see exactly where all the necessary files are stored for a specific project.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Tower MCP Server: 10 Tools for Project Management
Use these ten specific tools to run project operations—from creating tickets to listing members—all through your AI client's natural language commands.
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 Tower on Vinkius019d8490create task
Generates a brand new task item within a specified project.
019d8490get project
Retrieves detailed information about a single, specific project workspace.
019d8490get task details
Pulls all the specifics—assignee, description, status—for one existing task item.
019d8490list discussions
Fetches a list of recent topics and conversations for an entire project.
019d8490list doc folders
Lists all document folders associated with a given project, helping you find shared files.
019d8490list members
Retrieves a complete roster of team members available for assignment and collaboration.
019d8490list projects
Gets a list of every single project workspace you're connected to in Tower.
019d8490list tasks
Returns a summary list of all tasks currently active within a specified project.
019d8490list teams
Displays the names and structure of available team groupings in your workspace.
019d8490update task
Changes key details—like status, assignee, or description—on an existing task item.
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 Tower, then connect any of our 4,900+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,900+ 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 Tower. 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 server provides 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Tracking team progress shouldn't feel like clicking through five different dashboards.
Right now, if you need to know who owns a task or what the current status is, you open Tower. You navigate to the project, click the 'Tasks' tab, then filter by assignee, and finally scroll down to find that one item. It takes three clicks and maybe five seconds of scrolling.
With this MCP server, you just ask your agent: 'Show me all tasks assigned to John in Project X.' The agent runs `list_tasks` directly, bypassing the UI entirely. You get a clean, filtered list instantly.
Tower MCP Server helps you manage project workflows with natural language.
Manual updates require copy-pasting IDs, manually checking who is on the team via another system, and then updating the status in three different places to keep records clean. It's a mess of tabs and tribal knowledge.
Now you just tell your agent: 'Update Project Y with these details.' The agent calls `update_task` and handles the backend logic. You get confirmation; the data is accurate, everywhere.
What you can do with this MCP connector
Tower MCP Server - Manage Projects & Tasks Via AI
You've got your AI agent hooked up. It gives you direct command-line access to Tower, letting you treat complex project tracking like a conversation. You never gotta touch the web UI again. Your agent handles everything: listing projects, assigning tasks, checking status reports, and pulling shared files—all through simple natural language commands.
Discovering the Scope of Work
You wanna know what's going on across your whole portfolio? Run list_projects to get a real-time roster showing every collaboration workspace you have access to. Once you pick one, use get_project to pull all the deep details about that single project's setup and structure.
Controlling Tasks and Workflows
The core of this thing is task management. You can generate a brand new work item anytime by calling create_task, specifying exactly what the job is for within a certain project. Need to change something? Use update_task to instantly modify key details like changing status, adjusting the assignee, or rewriting the description on an existing ticket.
If you need the full picture of one specific task—who owns it, its current status, and its detailed notes—just run get_task_details. For a quick overview, however, list_tasks gives you a summarized list of every open item in that project.
Team Coordination and Resources
Managing people is just as important. Use list_members to pull up a complete roster of everyone on the team so you can make accurate assignments. You can also run list_teams to see the official names and structure of all available groupings in your workspace. When it comes to files, don't waste time searching folders; simply call list_doc_folders to get an immediate list of every shared resource folder tied to a project.
Monitoring Communication and Status
Keeping everyone aligned is key. To see what folks are talking about without reading hundreds of emails, run list_discussions. This pulls up the latest topics and conversations happening across the entire project history. You can combine this with running list_tasks to get a status summary, then use get_task_details if you need to know who's actually responsible for that specific item.
How It All Connects
Your agent doesn't just run tools; it connects them. You can pull the full list of projects via list_projects, then select one, and immediately get all available document locations using list_doc_folders. Need to assign someone? First, you check out the roster with list_members or see which teams are active with list_teams.
Once that's done, you can use those details when calling create_task to make sure the job lands on the right person. You gotta keep this flow going: list projects -> pick a project -> check members/files -> create task -> update status as it moves along.
It's pure command-line efficiency for managing everything from initial setup to final sign-off.
019d8490-f30c-715e-b1e3-3c1def0991b1 
How Tower MCP Works
- 1 Subscribe to this server and provide your Tower Access Token.
- 2 Connect your AI client (Claude, Cursor, etc.) to the Vinkius MCP endpoint.
- 3 Ask your agent a direct question: 'What tasks need updating in Project X?' The agent calls
list_tasksorget_project, gets the data, and answers you.
The bottom line is that your AI client talks directly to Tower's API tools, getting project data without needing a graphical interface.
Who Is Tower MCP For?
Project Managers who are tired of jumping between Jira and Slack. Operations Leads who need visibility across multiple workflows instantly. Creative Team Directors who spend too much time tracking down the latest discussion notes. If your job involves coordinating tasks across different people, this is for you.
Running status checks: 'Show me all overdue tasks in the API Integration project.' You use list_tasks and get_task_details to audit progress instantly.
Oversight tasks: 'List all teams and projects that require a resource review.' This uses list_teams, list_projects, and checks team structure.
Coordination tasks: 'Who is assigned to the logo design, and what did we discuss about it yesterday?' You chain calls using list_members and list_discussions.
What Changes When You Connect
- Skip the UI clicks. You don't have to navigate through five tabs to check a status. Just ask your agent, and it runs
list_tasksfor you right away. - Keep communication centralized. Use
list_discussionsto instantly pull up topic threads from any project, so you never miss an important decision made off-platform. - Assign work in natural language. Need a task created? Just tell your agent; it runs
create_task, assigning the item and logging it immediately. - See who's available. Run
list_membersto get all team names, then use that list when you need to update an assignment withupdate_task. - Audit project scope easily. You can run
get_projectto grab the full context and know exactly what boundaries a collaboration workspace operates within.
Real-World Use Cases
The Weekly Status Report
A PM needs to compile status for 10 projects. Instead of opening ten dashboards, they tell their agent: 'Give me the completion percentage and assignees for all tasks in Project X.' The agent runs list_tasks and gathers the data into one summary.
The Missing File Hunt
A developer needs the latest wireframes. Instead of guessing which folder it's in, they ask their AI client to 'Show me all document folders for Design Refresh.' The agent calls list_doc_folders and hands them the exact path.
Reassigning Ownership
A task was assigned incorrectly. Instead of manually editing the ticket, the ops lead tells their agent: 'Update Task ABC to assign it to Jane.' The agent executes update_task and confirms the change.
Onboarding New Team Members
A new hire joins the team. The manager asks the agent, 'Who are all the current members on the API Integration project?' The agent runs list_members, giving the manager a fresh roster for immediate assignment.
The Tradeoffs
Manual Project Audits
Copying and pasting task names from one spreadsheet into another because the PM needs to manually track progress across several unconnected tools.
→
Don't copy/paste. Use your agent to run list_projects, then drill down with list_tasks for specific data points, keeping everything within Tower.
Ambiguous Status Updates
Saying 'Update the task.' without specifying which task or what to update. This leads to failed calls and wasted time.
→
Be specific. Say: 'Run update_task for Project Alpha, changing the status of Task 1234 to Complete'—the agent knows exactly what to do.
Ignoring Dependencies
Trying to write a new task without checking if related discussions have been finalized first. You end up creating orphaned tickets.
→
Always check the context first. Run list_discussions for the project before you call create_task. It keeps your work connected and accurate.
When It Fits, When It Doesn't
Use this server if managing structured, collaborative data is your core job function. If you need to know: 'What projects exist?' (list_projects), 'Who is working on it?' (list_members), or 'What's the status of X?' (get_task_details), then Tower works for you.
Don't use this if your primary data source is unstructured content, like raw text documents or external databases that aren't housed in Tower. For those cases, look into a general database connector or file indexing solution instead. This tool lives and breathes within the Project Management paradigm.
Common Questions About Tower MCP
Can I use list_projects to find a project that doesn't exist in Tower? +
No. list_projects only retrieves projects you are actually connected to within the Tower workspace. It acts as your roster, not a search engine for outside data.
Does update_task require me to know the task ID? +
Yes. To successfully run update_task, you must provide the unique identifier of the existing task item. The agent can help you find this first using list_tasks.
What if I want to create a task but don't know the project ID? +
First, run list_projects to get all available IDs and names. Then, include the correct Project ID when you call create_task. You gotta provide that context.
How do I see who is on a team using list_members? +
You simply ask your agent to run list_members. It returns a full roster of names and roles, letting you know exactly who's available for assignments.
If I give an invalid access token when running list_projects, what error does the server return? +
The connection fails immediately. The agent will receive a specific 401 or authentication failure code from Vinkius before any project data is returned. You need to verify your Tower Access Token in the server settings.
When I call get_task_details, how do I know if a task was deleted versus just marked as complete? +
The tool response includes an explicit 'status' field. If the status is 'archived' or 'deleted', you won't be able to use update_task on it. You’ll still get all the original metadata.
If I run list_tasks and there are thousands of records, does the tool handle pagination automatically? +
Yes, your agent client handles pagination for large datasets. The response payload will contain a 'next page' token or total record count, allowing your agent to loop through all results until it hits the full dataset size.
Can I filter the output of list_tasks beyond just listing them? For instance, by assignee? +
You can. You pass a specific 'filter' parameter into the tool call—like an email address or user ID—to limit the result set before it reaches your agent. This keeps the data highly focused.
How do I obtain a Tower Access Token? +
Log in to your Tower account, go to the App Center, and generate a new Personal Access Token. Make sure to copy it immediately.
Can I see team discussions through this server? +
Yes. Use the list_discussions tool with a project ID to retrieve the list of recent topics and discussions shared within that project.
Is it possible to manage multiple teams? +
Yes. You can use list_teams to see all teams you are a part of and then use the team IDs to list members or access specific projects associated with those teams.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.