Alfred 5 Workflows MCP (macOS)

Alfred 5 Workflows MCP banner

For the macOS Alfred 5 app (alfredapp.com) with Powerpack — control your workflows from Claude, Cursor, and other MCP clients.
Not affiliated with the Alfred API platform (lumberjack-so/alfred-mcp).

Alfred 5 Workflows MCP connects AI assistants directly to macOS Alfred 5. It allows AI models to search, create, modify, execute, and package Alfred workflows using natural language.

Features

List & Search Workflows: Instantly scan installed Alfred workflows by name, bundle ID, or keyword.
Deep Inspection: Read workflow configurations, node objects, script code, plist data, and included files.
Create Workflows: Generate new Alfred 5 workflows from templates (keyword_script, script_filter, hotkey_trigger, empty) complete with UUIDs, nodes, and canvas positioning.
Edit & Update: Modify metadata (name, description, bundle ID, README) and update script contents or keywords in real-time.
File Management: Create or update embedded scripts, Python modules, shell files, or config assets directly inside workflow folders.
Trigger Actions: Fire Alfred External Triggers programmatically via macOS AppleScript.
Export & Package: Package workflows into standard .alfredworkflow files for distribution.
Safe Deletion: Safely trash workflows by moving them to macOS Trash via send2trash.

Prerequisites

macOS 12.0 or later.
Alfred 5 with an active Powerpack license.
Python 3.10 or higher.

Quick Setup & Onboarding

Choose your preferred installation method:

Option A: 1-Click Automated Setup (Recommended)

Run the automated installer script to create the environment, install dependencies, and automatically register alfred-mcp into Claude Desktop, Cursor, and Antigravity IDE:

git clone https://github.com/saihgupr/alfred-mcp.git
cd alfred-mcp
./install.sh

Option B: Zero-Install Execution (`uvx` / `pipx`)

If you use uv, you can run alfred-mcp directly without cloning or managing virtual environments manually. Add the following entry to your MCP client config (e.g. claude_desktop_config.json):

{
  "mcpServers": {
    "alfred": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/saihgupr/alfred-mcp.git", "alfred-mcp"]
    }
  }
}

Option C: Manual Setup

Clone & setup environment:

git clone https://github.com/saihgupr/alfred-mcp.git
cd alfred-mcp
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Run auto-installer command:

python server.py --install

or manually add the block below to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "alfred": {
      "command": "/path/to/alfred-mcp/.venv/bin/python",
      "args": ["/path/to/alfred-mcp/server.py"]
    }
  }
}

Available MCP Tools

Tool Name	Description
`alfred_list_workflows`	List all installed Alfred workflows with optional search query filtering (matches name, bundle ID, keyword).
`alfred_get_workflow`	Retrieve comprehensive workflow details including metadata, node objects, script code, connections, and files.
`alfred_create_workflow`	Create a new Alfred 5 workflow with proper `info.plist` structure, objects, and canvas connections.
`alfred_update_workflow_info`	Update metadata settings in a workflow's `info.plist` (name, description, bundle ID, README, disabled state, version).
`alfred_update_script_node`	Update the script code or keyword of a target script node inside a workflow.
`alfred_edit_file`	Create or overwrite files (e.g. `main.py`, `script.sh`, `config.json`) inside a workflow folder.
`alfred_delete_workflow`	Safely delete an Alfred workflow by moving its folder to macOS Trash.
`alfred_export_workflow`	Package a workflow directory into an installable `.alfredworkflow` file in Downloads or a specified folder.
`alfred_run_trigger`	Trigger an Alfred workflow's External Trigger using macOS AppleScript (`osascript`).

Troubleshooting & Debugging

If you encounter any issues while using alfred-mcp:

1. Log Locations

Claude Desktop Logs: Check ~/Library/Logs/Claude/mcp.log and ~/Library/Logs/Claude/mcp-server-alfred.log for execution errors.
Antigravity IDE Logs: Inspect ~/.gemini/antigravity-ide/mcp_config.json and agent session logs.
Cursor Logs: Open the Output panel in Cursor and select MCP Server Log.

2. macOS Automation Permissions

alfred_run_trigger uses AppleScript (osascript) to talk to Alfred 5.
Ensure your MCP client (Antigravity IDE / Claude Desktop / Cursor / Terminal) has permission under:
System Settings > Privacy & Security > Automation > [Your Client] > Alfred 5.

3. Alfred Preferences Sync Folder

alfred-mcp automatically scans default locations (~/Library/Application Support/Alfred/Alfred.alfredpreferences/workflows, Dropbox, and iCloud Sync paths).
If your workflows are stored in a custom folder, verify that Alfred 5 preferences plist (com.runningwithcrayons.Alfred-Preferences-3.plist) points to your current sync folder.

4. Interactive MCP Inspector

Test server tools locally with the MCP Inspector tool:

npx @modelcontextprotocol/inspector .venv/bin/python server.py

Contributing

Contributions are welcome! Please submit all Pull Requests to the main branch.

Fork the repository
Create a feature branch (git checkout -b feature/awesome-feature)
Commit your changes and push to your fork
Open a Pull Request to main

Support & Feedback

If you encounter any issues, bugs, or have feature requests, please open an issue on GitHub.

Alfred 5 Workflows MCP is open-source and free. If you find it useful, consider giving it a star ⭐ or making a donation to support development!