---
{
  "title": "Workflowy CLI",
  "intro": "Use Workflowy from your terminal, scripts, and AI agents."
}
---

The Workflowy CLI (`wf`) is an open-source command-line toolkit for reading, searching, organizing, and editing your Workflowy account. It includes a local cache for fast access, automation commands, machine-readable output for scripts and agents, and a built-in MCP server.

You can find the source code and latest releases in the [Workflowy CLI GitHub repository](https://github.com/rodolfo-terriquez/workflowy-cli).

# What you can do

- Read and search your Workflowy account from a terminal
- Add, update, move, complete, and delete nodes
- Work with built-in locations such as `@inbox`, `@today`, and `@tomorrow`
- Use bookmarks and paths to target specific parts of your account
- Export outlines, find todos and tags, and review recent changes
- Connect multiple Workflowy accounts and keep a separate local cache for each one
- Return structured JSON for scripts and AI agents
- Connect MCP-compatible AI tools directly to Workflowy

# What you need

- A Workflowy API key from [workflowy.com/api-key](https://workflowy.com/api-key/)
- A terminal on macOS, Linux, or Windows

# How to install

## macOS and Linux

Copy and paste this command into your terminal:

```bash
curl -fsSL https://github.com/rodolfo-terriquez/workflowy-cli/releases/latest/download/install.sh | bash
```

## Windows

Copy and paste this command into PowerShell:

```powershell
irm https://github.com/rodolfo-terriquez/workflowy-cli/releases/latest/download/install.ps1 | iex
```

The installer detects your operating system and processor, downloads the latest prebuilt version, and adds the `wf` command to your PATH when possible.

# Connect your Workflowy account

Run these commands after installation:

```bash
wf login
wf cache:sync
wf doctor
```

`wf login` asks for your Workflowy API key without displaying it. `wf cache:sync` downloads a local copy of your outline for fast reads and searches. `wf doctor` checks your login, configuration, cache, and installation.

# Start using the CLI

Read a location or search your account:

```bash
wf node:read @inbox
wf node:read @today --depth 3
wf search "quarterly planning"
wf targets
```

Add and manage items:

```bash
wf node:add @inbox "Draft project brief" --type todo
wf node:complete "@inbox/Draft project brief"
wf node:move "@today/Review roadmap" @inbox
wf node:update "@inbox/Follow up" --note "Waiting on a reply"
```

Commands that accept a target can use a built-in location, a bookmark returned by `wf targets`, a node ID, or a path such as `@today/Meetings/Launch review`.

Run `wf --help` to see the full command list, or add `--help` after a command for its options.

# Keep local results up to date

Most read and search commands use the local SQLite cache automatically. Sync it whenever you want the latest copy of your Workflowy account:

```bash
wf cache:sync
```

For a one-time request that should bypass the cache, use `--live` on supported commands:

```bash
wf node:read @inbox --live
wf search "launch notes" --live
```

# Connect multiple accounts

Give each account a short local name when you log in:

```bash
wf login --account personal
wf login --account work
wf account:list
wf account:switch work
wf account:current
```

Each account keeps its own local cache. You can select an account for a single command without changing your default:

```bash
wf --account personal cache:sync
wf --account work node:add @inbox "Follow up with the team"
```

# Use the CLI with scripts and AI agents

Add `--agent` to return machine-readable JSON with structured results and errors:

```bash
wf search "project roadmap" --agent
wf node:read @today --depth 3 --agent
```

The CLI also includes a local MCP server:

```bash
wf mcp
```

Configure your MCP client to run the installed `wf` binary with the argument `mcp`. Run `wf login` and `wf cache:sync` first so the server can use the same account and cache as the CLI. See the [Workflowy CLI README](https://github.com/rodolfo-terriquez/workflowy-cli#readme) for configuration examples and advanced MCP options.

# Update the CLI

Check for or install a newer release with:

```bash
wf self:update --check
wf self:update
```

# Security and troubleshooting

- Use the interactive `wf login` prompt so your API key is not saved in shell history.
- If search results look outdated, run `wf cache:sync` or use `--live`.
- If a command uses the wrong account, check `wf account:current` and switch with `wf account:switch <name>`.
- If `wf` is not found after installation, open a new terminal and check that the install directory is on your PATH.
- Run `wf doctor` for installation, authentication, configuration, and cache checks.
- For bugs and feature requests, visit the repository's [GitHub Issues](https://github.com/rodolfo-terriquez/workflowy-cli/issues).