MCP server for running AppleScript and JavaScript for Automation on macOS, with a searchable knowledge base of automation recipes for apps, files, browsers, system settings, Terminal, and productivity workflows.
macOS Automator MCP executes AppleScript or JavaScript for Automation through `osascript`, either from inline script content, an absolute script path, or a knowledge-base script ID., AppleScript and JXA can control local applications, read and write files, run shell commands through scripting additions, manipulate browser tabs, send keystrokes, change settings, and trigger app-specific actions depending on macOS permissions., Do not grant broad Automation or Accessibility permissions to the host process unless you trust the MCP client, prompts, and scripts it will run., Review generated scripts before execution, use short timeouts for untrusted automations, and avoid running scripts against Mail, Messages, Contacts, Calendar, browsers, password managers, terminals, or cloud-sync folders without explicit approval., Knowledge-base scripts and custom local knowledge-base overrides can change the behavior of reusable script IDs; review local overrides before relying on them., Use separate macOS user accounts or test machines when experimenting with destructive file, shell, browser, or system-setting automation.
Privacy notes
Scripts can expose local file paths, file contents, clipboard data, browser URLs and page data, email subjects, contacts, calendar events, messages, app state, system settings, shell output, and account-specific metadata., Error responses and optional debugging settings can include attempted script content, script paths, substitution logs, stderr, and execution details., The local knowledge base defaults to `~/.macos-automator/knowledge_base` when configured and may contain private automation recipes or paths., MCP clients, model providers, terminal history, and application logs may retain script inputs and outputs, including data read from local apps.
Author
Peter Steinberger
Submitted by
oktofeesh1
Claim status
unclaimed
Last verified
2026-06-06
Decision playbook
Review trust signals before you adopt
Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.
Compare context
Selected
0
Current score
63
Baseline
—
Delta
No baseline selected
No major trust-signal divergence detected in the current selection.
Source and provenance checks
Needs review
Confirm ownership and provenance before trusting install instructions.
Source link availableRequired
Open the canonical repository and verify ownership.
Done
Source provenance statusRequired
Marked as source-backed.
Done
Metadata reviewed
No reviewed flag detected in metadata.
Pending
Safety and privacy checks
Complete
Validate risk disclosures before installation or API wiring.
Safety notes presentRequired
Review the listed safety guidance before running commands.
Done
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Done
Trust level risk gateRequired
Trust level does not block evaluation.
Done
Package and install checks
Needs review
Check package metadata and artifact integrity signals.
Install payload available
Install or copy payload is available for review.
Done
Package verification flag
No package verification flag provided.
Pending
Checksum metadata
No checksum provided for downloaded artifact.
Pending
Compare-driven decision checks
Needs review
Use compare context to validate trade-offs before adoption.
Compare tray has multiple entries
Add at least one more entry to compare trust differences.
6 safety and 4 privacy notes across 5 risk areas. Review closely: credentials & tokens, permissions & scopes, third-party handling.
5 areas
SafetyLocal filesmacOS Automator MCP executes AppleScript or JavaScript for Automation through `osascript`, either from inline script content, an absolute script path, or a knowledge-base script ID.
SafetyPermissions & scopesAppleScript and JXA can control local applications, read and write files, run shell commands through scripting additions, manipulate browser tabs, send keystrokes, change settings, and trigger app-specific actions depending on macOS permissions.
SafetyPermissions & scopesDo not grant broad Automation or Accessibility permissions to the host process unless you trust the MCP client, prompts, and scripts it will run.
SafetyCredentials & tokensReview generated scripts before execution, use short timeouts for untrusted automations, and avoid running scripts against Mail, Messages, Contacts, Calendar, browsers, password managers, terminals, or cloud-sync folders without explicit approval.
SafetyExecution & processesKnowledge-base scripts and custom local knowledge-base overrides can change the behavior of reusable script IDs; review local overrides before relying on them.
SafetyLocal filesUse separate macOS user accounts or test machines when experimenting with destructive file, shell, browser, or system-setting automation.
PrivacyLocal filesScripts can expose local file paths, file contents, clipboard data, browser URLs and page data, email subjects, contacts, calendar events, messages, app state, system settings, shell output, and account-specific metadata.
PrivacyLocal filesError responses and optional debugging settings can include attempted script content, script paths, substitution logs, stderr, and execution details.
PrivacyLocal filesThe local knowledge base defaults to `~/.macos-automator/knowledge_base` when configured and may contain private automation recipes or paths.
PrivacyThird-party handlingMCP clients, model providers, terminal history, and application logs may retain script inputs and outputs, including data read from local apps.
Disclosure: Open-source MIT-licensed MCP server published as the `@steipete/macos-automator-mcp` npm package.
Safety notes
macOS Automator MCP executes AppleScript or JavaScript for Automation through `osascript`, either from inline script content, an absolute script path, or a knowledge-base script ID.
AppleScript and JXA can control local applications, read and write files, run shell commands through scripting additions, manipulate browser tabs, send keystrokes, change settings, and trigger app-specific actions depending on macOS permissions.
Do not grant broad Automation or Accessibility permissions to the host process unless you trust the MCP client, prompts, and scripts it will run.
Review generated scripts before execution, use short timeouts for untrusted automations, and avoid running scripts against Mail, Messages, Contacts, Calendar, browsers, password managers, terminals, or cloud-sync folders without explicit approval.
Knowledge-base scripts and custom local knowledge-base overrides can change the behavior of reusable script IDs; review local overrides before relying on them.
Use separate macOS user accounts or test machines when experimenting with destructive file, shell, browser, or system-setting automation.
Privacy notes
Scripts can expose local file paths, file contents, clipboard data, browser URLs and page data, email subjects, contacts, calendar events, messages, app state, system settings, shell output, and account-specific metadata.
Error responses and optional debugging settings can include attempted script content, script paths, substitution logs, stderr, and execution details.
The local knowledge base defaults to `~/.macos-automator/knowledge_base` when configured and may contain private automation recipes or paths.
MCP clients, model providers, terminal history, and application logs may retain script inputs and outputs, including data read from local apps.
Prerequisites
macOS on the machine running the MCP server.
Node.js 24 or newer for the published npm package.
Automation permissions for the terminal, editor, or app that launches the MCP server.
Accessibility permissions when scripts need UI scripting, clicks, keystrokes, or System Events access.
A clear boundary for which apps, files, accounts, and workflows Claude is allowed to automate.
Schema details
Install type
cli
Troubleshooting
No
Source repository stats
Scope
Source repo
Collection metadata
Estimated setup
10 minutes
Difficulty
advanced
Tool listing metadata
Disclosure
Open-source MIT-licensed MCP server published as the `@steipete/macos-automator-mcp` npm package.
macOS Automator MCP lets Claude execute AppleScript and JavaScript for
Automation on macOS. It exposes an execute_script tool for inline scripts,
absolute script paths, or knowledge-base script IDs, plus a get_scripting_tips
tool for discovering reusable automation recipes.
Use it when Claude needs supervised access to macOS automation tasks such as
controlling apps, reading browser or Finder state, preparing scripts, searching
the built-in knowledge base, or running vetted scripts under the permissions of
the host application.
These sources were reviewed on 2026-06-06. Prefer the live repository,
README, npm registry metadata, license, package manifest, launcher script,
server entrypoint, script executor, input schemas, and knowledge-base service
for current installation and behavior details.
Features
Run AppleScript through osascript.
Run JavaScript for Automation by passing the JavaScript language option.
claude mcp add macos-automator -- npx -y --package @steipete/macos-automator-mcp macos-automator-mcp
After adding the server, grant the terminal, editor, or app that launches the
server only the macOS Automation and Accessibility permissions needed for the
apps and workflows you intend to automate.
Use Cases
Ask Claude to find an existing AppleScript recipe before writing a new one.
Run a vetted script that reads the current Safari tab URL.
Automate Finder file organization on a test folder.
Control app windows or settings under macOS Automation permissions.
Prototype JXA scripts with explicit timeouts and visible output.
Build a private local knowledge base of approved automation recipes.
Debug script permissions and output formatting for a macOS workflow.
Safety and Privacy
This server runs scripts on the local Mac. Treat execute_script like a local
automation and code-execution capability, not a read-only helper. Review scripts
before running them, keep permissions narrow, and test destructive or app-writing
workflows on disposable files or a separate macOS user account.
Automation and Accessibility permissions can allow the host process to control
apps, inspect UI state, send keystrokes, click buttons, read windows, and change
settings. Give those permissions only to trusted launchers and only for the apps
needed by the workflow.
Script output can include private local data from files, browser tabs,
clipboard, messages, email, calendars, contacts, shell commands, and app state.
Avoid sharing transcripts that include script inputs, outputs, paths, local
knowledge-base entries, or permission errors.
Duplicate Check
Existing entries include narrower macOS automation surfaces such as iTerm MCP,
but no macOS Automator MCP, steipete/macos-automator-mcp,
@steipete/macos-automator-mcp, or general AppleScript/JXA automation MCP entry
was found in content/mcp.
Show that macOS Automator MCP is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.
[](https://heyclau.de/entry/mcp/macos-automator-mcp)
How it compares
macOS Automator MCP side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
MCP server for running AppleScript and JavaScript for Automation on macOS, with a searchable knowledge base of automation recipes for apps, files, browsers, system settings, Terminal, and productivity workflows.
MCP server for Anki flashcards that lets Claude review cards, manage decks, create and update notes, handle tags and media, inspect stats, and drive Anki desktop GUI workflows.
MCP and REST server for creating short-form videos from scene text, Pexels search terms, text-to-speech, captions, background music, Whisper captions, FFmpeg, and Remotion rendering.
✓macOS Automator MCP executes AppleScript or JavaScript for Automation through `osascript`, either from inline script content, an absolute script path, or a knowledge-base script ID.
AppleScript and JXA can control local applications, read and write files, run shell commands through scripting additions, manipulate browser tabs, send keystrokes, change settings, and trigger app-specific actions depending on macOS permissions.
Do not grant broad Automation or Accessibility permissions to the host process unless you trust the MCP client, prompts, and scripts it will run.
Review generated scripts before execution, use short timeouts for untrusted automations, and avoid running scripts against Mail, Messages, Contacts, Calendar, browsers, password managers, terminals, or cloud-sync folders without explicit approval.
Knowledge-base scripts and custom local knowledge-base overrides can change the behavior of reusable script IDs; review local overrides before relying on them.
Use separate macOS user accounts or test machines when experimenting with destructive file, shell, browser, or system-setting automation.
✓AI Game Developer can create, move, copy, modify, and delete Unity assets, scenes, GameObjects, components, scripts, packages, and generated project files.
Tools include dynamic C# script execution, C# reflection method lookup and calls, package installation/removal, Unity test execution, editor state changes, play mode control, screenshots, and profiler access.
The MCP server supports streamable HTTP and stdio; HTTP deployments should require a bearer token and stay bound to trusted interfaces.
Server variables include optional authorization and webhook settings; webhook endpoints can receive tool, prompt, resource, connection, and authorization events.
Unity runtime connections can expose compiled game state or in-game behavior to an MCP client, not just editor-only project data.
Use source control, backups, tool filtering, and explicit review before allowing destructive tools such as asset deletion, package removal, script execution, or reflection calls.
✓Anki MCP Server can create, update, delete, move, tag, rate, and sync cards and notes through AnkiConnect.
Destructive tools include note deletion, media deletion, unused-tag clearing, deck moves, tag replacement, model styling changes, and review rating actions that affect scheduling.
The server supports read-only mode that blocks exposed write actions, but users should still review tool calls before enabling unrestricted study or collection-management workflows.
Media tools can store files from local paths, URLs, or base64 data and can retrieve media as base64 content.
GUI tools can open Anki windows, select cards, edit notes, show answers, and undo the last Anki action.
Remote tunnel or ngrok-style setups expose local Anki to cloud-hosted AI clients and require careful authentication, origin, and URL-sharing controls.
✓Short Video Maker runs an HTTP server with MCP SSE routes and REST routes for creating and serving generated videos.
The create-short-video tool queues video rendering work that can consume CPU, memory, disk, browser rendering resources, and optional GPU resources.
Generated videos combine user-provided narration text, Pexels search terms, downloaded background clips, local music assets, TTS output, captions, and rendered media.
The server stores generated videos and temporary files under its configured data directory.
Validate scene text, search terms, claims, branding, and rights before publishing generated videos to TikTok, Instagram, YouTube, or other platforms.
Privacy notes
✓Scripts can expose local file paths, file contents, clipboard data, browser URLs and page data, email subjects, contacts, calendar events, messages, app state, system settings, shell output, and account-specific metadata.
Error responses and optional debugging settings can include attempted script content, script paths, substitution logs, stderr, and execution details.
The local knowledge base defaults to `~/.macos-automator/knowledge_base` when configured and may contain private automation recipes or paths.
MCP clients, model providers, terminal history, and application logs may retain script inputs and outputs, including data read from local apps.
✓Tool calls may expose Unity project paths, asset names, scene hierarchy, serialized object data, scripts, logs, screenshots, profiler metrics, test output, package metadata, and runtime state.
MCP config files and Unity plugin config can contain server URLs, connection modes, bearer tokens, authorization settings, enabled tool IDs, and cloud or local endpoint details.
Optional webhooks can receive tool, prompt, resource, connection, authorization, and token-bearing request data.
Screenshots from Game View, Scene View, cameras, or isolated GameObjects may include proprietary artwork, level design, UI, debug overlays, or unreleased game content.
Generated skills and AI-client configuration files can reveal available tools, project structure, installed packages, and local workflow assumptions.
✓Tool calls can expose deck names, note fields, card contents, tags, review history, media filenames, base64 media, selected notes, collection stats, scheduling metadata, and AnkiConnect errors.
Flashcards may contain medical, language-learning, school, work, personal, customer, or confidential study material.
Sync can send Anki collection changes to AnkiWeb according to the user's Anki account and sync settings.
Local media file paths, URLs imported into cards, and GUI-selected note IDs can enter MCP client context and model logs.
Remote tunnel providers and hosted clients may process connection metadata or traffic according to their own policies.
✓`PEXELS_API_KEY` is sent to the Pexels API when searching for background videos.
Scene text, search terms, generated narration, captions, video IDs, logs, output filenames, and render status can reveal campaign ideas, private scripts, brand plans, or client work.
Generated files remain on disk until removed according to the deployment's data-directory and retention practices.
Pexels requests, model downloads, Docker pulls, npm installs, and platform uploads may expose network metadata to third parties outside the MCP client.
Prerequisites
macOS on the machine running the MCP server.
Node.js 24 or newer for the published npm package.
Automation permissions for the terminal, editor, or app that launches the MCP server.
Accessibility permissions when scripts need UI scripting, clicks, keystrokes, or System Events access.
Unity project using Unity 2022.3 or newer for the current package metadata.
Unity Hub or a local Unity Editor installation.
Node.js 20.19 or newer, or Node.js 22.12 or newer, for `unity-mcp-cli`.
An MCP client such as Claude Code, Claude Desktop, Codex, Cursor, Gemini CLI, GitHub Copilot, Cline, or another supported client.
Anki Desktop installed and running.
The AnkiConnect add-on installed in Anki.
Node.js 22.12 or newer for the npm package path.
Review of which decks, cards, tags, media files, and note types the MCP client may access.
Node.js or Docker available on the machine that will run the video server.
Pexels API key configured through `PEXELS_API_KEY`.
Enough CPU, memory, and disk for Whisper, Kokoro, FFmpeg, Remotion, temporary files, and rendered videos.
MCP client that can connect to an SSE MCP endpoint.