--- layout: tap site_name: shortcuts tap_name: run description: "Run an Apple Shortcut" intent: read columns: - output args: - name: name type: string description: "Shortcut name" - name: input type: string description: "Input text (optional)" args_json: | {"name":{"type":"string","required":true,"description":"Shortcut name"},"input":{"type":"string","default":"","description":"Input text (optional)"}} health_json: | {"min_rows":1,"non_empty":[]} example_args: "" source_url: https://github.com/LeonTing1010/tap-skills/blob/main/community/shortcuts/run.plan.json license: MIT ---

What it does

Run an Apple Shortcut

Install Taprun once

Taprun ships as a single MCP server exposing a catalog of compiled taps. One-time setup on macOS / Linux:

brew install LeonTing1010/tap/taprun
tap mcp connect

Or drop this into your claude_desktop_config.json (works identically in Claude Code, Cursor, Cline, Windsurf — any MCP host):

{
  "mcpServers": {
    "tap": {
      "command": "tap",
      "args": ["mcp", "start"]
    }
  }
}

Call shortcuts/run

Terminal, once installed:

tap run shortcuts/run

From the MCP host — exact same compiled plan, deterministic replay, zero LLM tokens:

tap.run({ site: "shortcuts", name: "run" })

Why compile it once

This plan was forged once — the AI read shortcuts, picked stable structural addresses (JSON-LD, ARIA, RSS, or declared API endpoints, in that priority order), and saved them to a .plan.json. Every replay since then has used zero LLM tokens. When shortcuts ships a site change that breaks the extraction, tap verify surfaces it before your data goes stale — not after your pipeline silently writes garbage for a week.

Related shortcuts taps

tapdescription
shortcuts/listList available Apple Shortcuts