Guides
Straight answers.
What people ask before they lend an AI their Mac. Each answer starts short, then shows the detail.
How do I let Claude Code control my Mac?
Install Waldo, a free and open source MCP server for macOS, and register him with Claude Code. Run make install, then waldo grant for Accessibility and Screen Recording, then make register. Start a new Claude Code session and ask in plain words. Claude Code can then see every window, press every button, and type into every field. Money, messages, passwords, and deletes still stop and ask you.
Read the answerIs there an MCP server that lets an AI agent use my Mac?
Yes. Waldo is a free, open source MCP server for macOS, under Apache 2.0. Any agent that speaks MCP can use his tools to see every window, press every button, and type into every field, in apps in front or behind. He reads the accessibility tree, not just pixels, checks that each act worked, and stops to ask you before money, messages, passwords, or deletes.
Read the answerHow do I give Codex, Gemini CLI, or Cursor hands on my Mac?
Install Waldo, run waldo grant, then add one MCP server to your agent's settings: a server named waldo that runs the command waldo with the argument mcp. Codex, Gemini CLI, Cursor and friends all take the same command. macOS asks permission for whichever app starts it, so grant Accessibility and Screen Recording to that app.
Read the answerWhat is computer use for AI agents?
Computer use means an AI agent operates a computer the way a person does. It looks at the screen, then clicks, types, and presses keys in real apps. Many systems look at screenshots and click by pixel position. Waldo reads the accessibility tree instead, so each control comes with its name, role, and real position. Four kinds of act always stop and ask you first.
Read the answerHow do I stop an AI agent paying, sending, or deleting without me?
Put the stops in code, not in a prompt. Waldo, a free and open source MCP server for macOS, gates four kinds of act: money, sending as you, passwords and keys, and destructive changes. The code decides when a gate applies, from facts it reads off the screen. Then it asks you, and only a real click from your own mouse can say yes. The agent cannot give its own yes.
Read the answerShould AI agents read screenshots or the accessibility tree?
Lead with the accessibility tree, and keep screenshots for looking. The tree is the same one screen readers use. Every control arrives with its name, role, and real position, so the model picks a thing instead of guessing at pixels. Waldo reads all 241 controls in a Chrome window in 0.35 seconds. Screenshots still help when the model needs to see, and when an app draws its own controls and has no tree.
Read the answerHow do I stop an AI agent at once on my Mac?
Push your pointer into the top-left corner of the screen. With Waldo, every tool, for every agent, stops until you say go. You can also run waldo stop in a terminal, or press Stop everything in the Waldo bar or on a gate panel. Resume with waldo resume or the bar's Resume button.
Read the answer