Update, October 2026: Since this post, the MCP server has grown to around 40 tools — including your notification inbox, automation rules, wiki pages, and moving items between boards. Clients like Claude can now also connect via OAuth: enter the URL, log in, approve — no API key needed — and both API keys and OAuth connections can be limited to read-only. The MCP setup guide has the current steps.
Imagine asking your AI assistant: “What tasks are overdue in the Marketing workspace?” and getting a real answer — not a hallucination based on training data, but live data pulled directly from your project board. Or saying “Create a bug report for the login issue and assign it to Sarah with high priority” and having it done instantly. That’s what TskVlo’s MCP integration makes possible.
What Is MCP?
The Model Context Protocol (MCP) is an open standard developed by Anthropic that defines how AI language models can securely connect to external tools and data sources. Think of it as a universal API layer between your AI assistant and the rest of your digital world. Instead of copying and pasting information into a chat window, MCP lets the AI read from and write to external systems directly — with your permission.
MCP uses a client-server architecture. Your AI client (like Claude) connects to an MCP server, which exposes a set of named “tools” the AI can call. Each tool has a name, description, and typed parameters. The AI decides which tools to call based on your instructions, the server executes them, and the results flow back into the conversation.
TskVlo’s MCP Server
TskVlo includes a production-ready MCP server as a first-class feature — not an afterthought. It comes with every TskVlo workspace — there’s nothing to install or run. At launch it exposed around 25 tools covering every major operation:
- Board tools:
list_boards,create_board,update_board,delete_board - Group tools:
list_groups,create_group,update_group,delete_group - Item tools:
list_items,get_item,create_item,update_item,delete_item - Field tools:
list_board_fields,create_field,update_field,delete_field - Value tools:
set_item_field_value,set_item_field_values,clear_item_field_value - File tools:
list_item_files,upload_item_file,delete_item_file - Comment tools:
list_comments,create_comment
Every tool requires authentication — a workspace API key or, since then, an OAuth sign-in — so your data is never exposed without explicit authorization.
Connecting Claude to TskVlo
Here’s how to connect with an API key in four steps. (If your client supports OAuth, such as Claude’s custom connectors, it’s even simpler: add the URL below as a connector, log in, and click Authorize.)
Step 1: Generate an API Key
Log in to TskVlo as a workspace Owner or Admin, navigate to Workspace settings → API keys, and create a new key. Give it a descriptive name (e.g., “Claude MCP”) and copy the generated key — you won’t see it again.
Step 2: Find Your MCP Server URL
The TskVlo MCP server is available at:
https://dash.tskvlo.app/api/mcpStep 3: Configure Your MCP Client
In clients that take a JSON configuration with a URL and headers — Cursor’s mcp.json, for example — add the following:
{
"mcpServers": {
"tskvlo": {
"url": "https://dash.tskvlo.app/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY_HERE"
}
}
}
}Step 4: Test the Connection
Restart your client and ask something simple:
List all the boards in my TskVlo workspace.The AI will call the list_boards tool and return a formatted list of your boards. From there, you can start exploring more complex workflows.
Example Prompts That Actually Work
Once connected, here are some prompts you can try:
- “Show me all items in the ‘Website Redesign’ board that are marked as In Progress.”
- “Create a new board called ‘Q3 Marketing Campaign’ in the Marketing workspace with a Status field and a Due Date field.”
- “Move all tasks assigned to Alex that are past their due date to the ‘Blocked’ status.”
- “How many open bugs are there in the Development workspace?”
- “Create five sub-tasks under the ‘Launch Checklist’ item for SEO, analytics setup, social media, email blast, and press release.”
The AI handles the tool calls behind the scenes. You just describe what you want in plain language.
A Note on Security
TskVlo’s MCP server never reaches beyond the workspace it was connected to: every API key is bound to a single workspace, and OAuth connections have to be approved by a workspace Owner or Admin for one specific workspace. Both can be limited to read-only — the AI then doesn’t even see the write tools. We recommend a dedicated API key per integration, and read-only access wherever the AI only needs to look things up.
The future of project management isn’t just better UIs — it’s AI that understands your entire workflow context. TskVlo’s MCP integration is our first step toward that vision.