Skip to main content
The Arca MCP server enables AI assistants like Claude Desktop, Cline, and other MCP-compatible tools to access your Arca workspaces, tasks, lists, and folders directly. This allows you to manage your projects using natural language commands through your preferred AI assistant.

​
What is MCP?

Model Context Protocol (MCP) is an open standard developed by Anthropic that allows AI assistants to securely connect to external data sources and tools. The Arca MCP server implements this protocol to bridge your AI assistant with your Arca account.

​
What can you do with the MCP server?

Once connected, your AI assistant can:
  • List and search workspaces, tasks, lists, and folders
  • Create new tasks with titles, descriptions, due dates, and priorities
  • Update existing tasks - change status, priority, assignees, dates
  • Organize work - create and manage lists and folders
  • Add context - post comments on tasks
  • Delete items - remove tasks, lists, or folders when needed
All actions are performed using your Arca API key, ensuring everything is tracked and attributed to your account.

​
Installation

The MCP server is distributed via npm and can be installed instantly using npx: 
npx arca-mcp
No manual installation is needed! The npx command automatically downloads and runs the latest version.

​
Get your API key

Before configuring the MCP server, you need an Arca API key.
1

Open Arca Settings

Click your avatar in the bottom-left sidebar, then select Settings.
2

Navigate to API Keys

In the settings sidebar, click API Keys under the Developer section.
3

Create a new key

Click Create API Key, give it a descriptive name (e.g., “Claude Desktop”), choose all permissions needed, and click Create.
4

Copy your key

Copy the generated API key. It starts with arca_ and you won’t be able to see it again once you close the dialog.
Keep your API key secure! Anyone with this key can access your Arca workspaces. Never share it publicly or commit it to version control.

​
Configuration

​
Claude Desktop

Claude Desktop has native MCP support. To connect it to Arca:
1

Locate the config file

Open your Claude Desktop configuration file:macOS: ~/Library/Application Support/Claude/claude_desktop_config.jsonWindows: %APPDATA%\Claude\claude_desktop_config.jsonLinux: ~/.config/Claude/claude_desktop_config.json
2

Add the Arca MCP server

Add the following configuration to the mcpServers section:
{
  "mcpServers": {
    "arca": {
      "command": "npx",
      "args": ["-y", "arca-mcp"],
      "env": {
        "ARCA_API_KEY": "arca_your_api_key_here"
      }
    }
  }
}
Replace arca_your_api_key_here with your actual API key from the previous step.
3

Restart Claude Desktop

Quit Claude Desktop completely and reopen it. The MCP server will be loaded automatically.
4

Verify the connection

In Claude Desktop, click the 🔌 plug icon in the bottom-right corner. You should see “arca” listed with a green indicator showing it’s connected.

​
Cline (VS Code Extension)

Cline is an AI coding assistant that runs in VS Code and supports MCP.
1

Open Cline settings

In VS Code, open the Cline extension settings (usually via the gear icon in the Cline panel).
2

Add MCP server

Navigate to the MCP Servers section and add a new server with: - Name: arca - Command: npx - Args: ["-y", "arca-mcp"] - Environment Variables: ARCA_API_KEY=arca_your_api_key_here
3

Reload Cline

Restart the Cline extension or reload VS Code. The Arca MCP server will be available in your next chat.

​
Other MCP-compatible clients

Any tool that supports the Model Context Protocol can connect to the Arca MCP server. The general pattern is:
  1. Command: npx
  2. Arguments: ["-y", "arca-mcp"]
  3. Environment variable: ARCA_API_KEY with your API key
Refer to your AI assistant’s documentation for specific MCP configuration instructions.

​
Example commands

Once connected, you can use natural language to interact with Arca:
You: “Show me all my Arca workspaces” Assistant: Lists all workspaces with their IDs, names, slugs, and member counts
You: “Create a task in my workspace called ‘Fix login bug’ with high priority and due date next Friday” Assistant: Creates the task with the specified details and returns the task ID
You: “Mark task #125 as completed” Assistant: Updates the task status and confirms the change
You: “Show me all high-priority tasks in workspace 1 that are assigned to me” Assistant: Filters and displays matching tasks
You: “Add a comment to task #89 saying ‘Reviewed and approved’” Assistant: Posts the comment and confirms
You: “Create a new list called ‘Q2 Goals’ in my Marketing workspace” Assistant: Creates the list and returns its ID

​
Available tools

The MCP server provides 20 tools for interacting with Arca:
ToolDescription
list_workspacesList all workspaces you have access to
get_workspaceGet details of a specific workspace
list_tasksList tasks in a workspace (with optional filters)
get_taskGet full details of a specific task
create_taskCreate a new task
update_taskUpdate an existing task
delete_taskDelete a task
list_listsList all lists in a workspace
create_listCreate a new list
update_listUpdate an existing list
delete_listDelete a list
list_foldersList all folders in a workspace
create_folderCreate a new folder
update_folderUpdate an existing folder
delete_folderDelete a folder
list_commentsList all comments on a task
create_commentAdd a comment to a task

​
Troubleshooting

Possible causes: - Invalid API key - verify it starts with arca_ and is copied correctly - Network connectivity issues - check your internet connection - Outdated npm cache - run npx clear-npx-cache and try again Solution: Double-check your API key in the config file and restart your AI assistant.
Possible causes: - API key belongs to a user without workspace access - API key was revoked or expired Solution: Verify your API key is still valid in Arca Settings → API Keys. Generate a new key if needed.
Possible causes: - Trying to modify a workspace where you don’t have edit permissions - Attempting to delete items without proper role Solution: Check your workspace role. Viewers can only read data, while Members and Admins can create and modify tasks.
Claude Desktop: - Make sure you’re editing the correct config file path for your OS - Create the file if it doesn’t exist - Ensure it’s valid JSON (use a JSON validator) Solution: Create the directory if missing:
  # macOS/Linux mkdir -p ~/Library/Application\ Support/Claude # Windows
  PowerShell New-Item -ItemType Directory -Force -Path "$env:APPDATA\Claude"
Cause: Node.js is not installed or not in your PATH. Solution: Install Node.js from nodejs.org. Version 16 or higher is required.

​
Security & privacy

  • All API requests are made directly from your machine to Arca’s servers
  • Your API key is stored locally in your AI assistant’s config file
  • The MCP server never stores or transmits your data to third parties
  • API keys can be revoked anytime from Arca Settings → API Keys

​
Updates

The MCP server is automatically updated when you use npx arca-mcp. To ensure you have the latest features and bug fixes, restart your AI assistant periodically. You can check the current version: 
npm view arca-mcp version

​
Need help?