Workflowy

Exposes read and write Workflowy actions via an MCP server for AI assistants
  • go

7

GitHub Stars

go

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "mholzen-workflowy": {
      "command": "workflowy",
      "args": [
        "mcp",
        "--expose=all",
        "--log-file=/tmp/workflowy-mcp.log"
      ],
      "env": {
        "WORKFLOWY_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

You can run Workflowy as an MCP server to expose its node operations to AI assistants. This enables read and write interactions with your Workflowy data through structured tools, automations, and chat-based workflows while keeping control over access through key-based authentication and logging.

How to use

Configure an MCP client to connect to Workflowy using the available stdio MCP endpoints. You’ll load the server configuration, provide your API key, and start the MCP server so your client can access read and write tools. You can run as read-only or enable full access as needed. Use the read-only endpoint first to verify connectivity, then enable all tools if you require write operations.

How to install

Install prerequisites and set up the MCP server entry points as shown. You will use a local CLI binary to expose MCP endpoints to your client.

# Install the Workflowy CLI that includes MCP support
# (example: install from your package manager or binary as applicable in your setup)
# Ensure you have a working Go or Node environment if required by your install flow

Configuration and usage notes

Enable MCP by configuring a local stdio server entry that launches the Workflowy CLI with the MCP mode. You can also run a read-only version for safer access. Provide your API key to authenticate requests.

Examples

The following configurations expose the MCP endpoints for Workflowy. One enables all tools (read and write), and the other runs in read-only mode.

{
  "mcpServers": [
    {
      "name": "workflowy",
      "type": "stdio",
      "command": "workflowy",
      "args": ["mcp", "--expose=all", "--log-file=/tmp/workflowy-mcp.log"],
      "env": [
        {"name": "WORKFLOWY_API_KEY", "value": "YOUR_API_KEY"}
      ]
    },
    {
      "name": "workflowy_readonly",
      "type": "stdio",
      "command": "workflowy",
      "args": ["mcp", "--log-file=/tmp/workflowy-mcp.log"],
      "env": [
        {"name": "WORKFLOWY_API_KEY", "value": "YOUR_API_KEY"}
      ]
    }
  ]
}

"],"language":

## Security and access control

Keep your API key secure. Use the read-only MCP endpoint for non-destructive tasks and only upgrade to full access when you trust the integration. Log files help audit usage and diagnose issues.

## Troubleshooting

If the MCP client cannot connect, verify that the Workflowy CLI is installed, the API key is valid, and the log file path is writable. Check for permission errors and ensure the MCP process has access to the key and log paths.

## Notes

All available tools are exposed through the MCP server as read and write endpoints. Start with read-only access to confirm connectivity, then enable write capabilities if you need to create, update, or delete nodes.

## Examples of common workflows

- Retrieve a node and its descendants via the MCP get-style tool
- List descendants and feed results into an automation
- Search for items by text or regex and process matches in a bot
- Create, update, or delete nodes through write-enabled MCP tools once authorized

## Available tools

### workflowy\_get

Get a node and its descendants, with options to specify depth and method for API access.

### workflowy\_list

List descendants as a flat list, with options for depth and all-descendants mode.

### workflowy\_search

Search nodes by text or regex with optional item scope and format outputs.

### workflowy\_targets

List available targets like inbox and shortcuts for use as IDs in commands.

### workflowy\_report\_count

Generate a descendant count report and upload or print results.

### workflowy\_report\_children

Rank nodes by their number of immediate children.

### workflowy\_report\_created

Rank nodes by creation date to find oldest items.

### workflowy\_report\_modified

Rank nodes by last modification date to find stale content.

### workflowy\_create

Create a new node under a specified parent.

### workflowy\_update

Update an existing node's content or metadata.

### workflowy\_delete

Delete a node permanently.

### workflowy\_complete

Mark a node as complete.

### workflowy\_uncomplete

Mark a node as not complete.

### workflowy\_replace

Bulk find-and-replace text in node names with regex support.
Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational