safari-mcp 2.7.10 → 2.7.12

package/README.md

@@ -13,6 +13,7 @@
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 [![macOS](https://img.shields.io/badge/macOS-only-blue)](https://www.apple.com/macos/)
 [![GitHub stars](https://img.shields.io/github/stars/achiya-automation/safari-mcp?style=social)](https://github.com/achiya-automation/safari-mcp/stargazers)
+[![Featured in CLI-Anything](https://img.shields.io/badge/Featured%20in-CLI--Anything-blueviolet)](https://github.com/HKUDS/CLI-Anything/pull/212)
 
 <a href="vscode:mcp/install?%7B%22safari-mcp%22%3A%7B%22command%22%3A%22npx%22%2C%22args%22%3A%5B%22safari-mcp%22%5D%7D%7D"><img src="https://img.shields.io/badge/VS_Code-Install_Server-0078d4?logo=visual-studio-code&logoColor=white" alt="Install in VS Code"></a>
 <a href="https://insiders.vscode.dev/redirect?url=vscode-insiders:mcp/install?%7B%22safari-mcp%22%3A%7B%22command%22%3A%22npx%22%2C%22args%22%3A%5B%22safari-mcp%22%5D%7D%7D"><img src="https://img.shields.io/badge/VS_Code_Insiders-Install_Server-24bfa5?logo=visual-studio-code&logoColor=white" alt="Install in VS Code Insiders"></a>

package/index.js

@@ -1009,6 +1009,33 @@ server.tool(
   }
 );
 
+server.tool(
+  "safari_native_keyboard",
+  "OS-level keyboard event via macOS CGEvent — sends a real keypress (with optional modifiers) to the Safari window WITHOUT activating Safari or stealing focus. Use when safari_press_key's JS path doesn't reach React trust-gated handlers (Discord ProseMirror Enter, Slack send, virtualized editors). Keys: enter, return, tab, escape, space, delete, backspace, up/down/left/right, home, end, pageup, pagedown, f1-f6, a-z, 0-9 and common punctuation. Modifiers: cmd, shift, alt, ctrl. Produces isTrusted:true events. Never activates Safari — runs entirely in the background.",
+  {
+    key: z.string().describe("Key name: enter, escape, tab, space, arrow keys, letters, digits, etc."),
+    modifiers: z.array(z.string()).optional().default([]).describe("Modifier keys: cmd, shift, alt, ctrl"),
+  },
+  async (args) => {
+    const result = await safari.nativeKeyboard(args);
+    return { content: [{ type: "text", text: typeof result === 'string' ? result : JSON.stringify(result) }] };
+  }
+);
+
+server.tool(
+  "safari_native_type",
+  "Insert text into ANY editor via OS-level clipboard paste (CGEvent Cmd+V targeted to Safari window). Unlike safari_fill which manipulates DOM directly (breaking React/ProseMirror state), this goes through the real paste pipeline — ProseMirror/Slate/Draft.js process the paste event natively and update their internal model. After native_type, pressing Enter (via safari_native_keyboard) will actually submit the form because the framework state matches the DOM. Saves and restores the user's clipboard. No focus stealing. Use for Discord, Slack, and any editor where safari_fill works visually but the content isn't 'really there' when you try to submit.",
+  {
+    value: z.string().describe("Text to insert via clipboard paste"),
+    selector: z.string().optional().describe("CSS selector of the editor element to focus first"),
+    ref: z.string().optional().describe("Ref ID from safari_snapshot to focus first"),
+  },
+  async (args) => {
+    const result = await safari.nativeType(args);
+    return { content: [{ type: "text", text: typeof result === 'string' ? result : JSON.stringify(result) }] };
+  }
+);
+
 // ========== FORM INPUT ==========
 
 server.tool(

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "safari-mcp",
-  "version": "2.7.10",
+  "version": "2.7.12",
   "mcpName": "io.github.achiya-automation/safari-mcp",
   "description": "Safari browser automation for AI agents — native macOS, zero Chrome overhead. 80 tools via AppleScript + JavaScript.",
   "type": "module",

package/safari.js

@@ -1397,6 +1397,62 @@ const jsKeyMap = {
   f1: "F1", f2: "F2", f3: "F3", f4: "F4", f5: "F5", f6: "F6",
 };
 
+// macOS virtual key codes for CGEvent keyboard (used by _helperNativeKeyboard).
+// These are the HID-level codes that postToPid uses, NOT JS keyCode values.
+const macKeyCodeMap = {
+  enter: 36, return: 36, "numpad-enter": 76,
+  tab: 48, space: 49, delete: 51, backspace: 51, escape: 53,
+  up: 126, "arrowup": 126, down: 125, "arrowdown": 125,
+  left: 123, "arrowleft": 123, right: 124, "arrowright": 124,
+  home: 115, end: 119, pageup: 116, pagedown: 121,
+  f1: 122, f2: 120, f3: 99, f4: 118, f5: 96, f6: 97,
+  a: 0, s: 1, d: 2, f: 3, h: 4, g: 5, z: 6, x: 7, c: 8, v: 9,
+  b: 11, q: 12, w: 13, e: 14, r: 15, y: 16, t: 17,
+  "1": 18, "2": 19, "3": 20, "4": 21, "6": 22, "5": 23,
+  "=": 24, "9": 25, "7": 26, "-": 27, "8": 28, "0": 29,
+  "]": 30, o: 31, u: 32, "[": 33, i: 34, p: 35,
+  l: 37, j: 38, "'": 39, k: 40, ";": 41, "\\": 42,
+  ",": 43, "/": 44, n: 45, m: 46, ".": 47, "`": 50,
+};
+
+// Native keyboard via CGEvent — sends a single key (with optional modifiers)
+// to the Safari window WITHOUT activating Safari or moving the mouse.
+// This produces isTrusted:true events that bypass React trust checks (Discord ProseMirror,
+// Slack virtualized editors, etc.) without any focus stealing. Requires Safari window ID.
+// Native type via clipboard paste (CGEvent Cmd+V targeted to Safari window).
+// Inserts text into ANY editor (ProseMirror, Slate, Draft.js, regular inputs,
+// contenteditable, cross-origin iframes) by going through the real paste pipeline.
+// Unlike safari_fill's synthetic paste, this updates the framework's internal
+// model state, so subsequent operations (like pressing Enter to submit in Discord)
+// see the text as "really there".
+export async function nativeType({ value, selector, ref }) {
+  await ensureHelpers();
+  if (!value) throw new Error("nativeType requires 'value'");
+  // Focus the target element if selector/ref provided
+  if (ref || selector) {
+    const sel = ref ? refSelector(ref) : selector.replace(/\\/g, "\\\\").replace(/'/g, "\\'");
+    await runJS(`(function(){ var el=document.querySelector('${sel}'); if(el){el.focus();el.click();} return el?'focused':'not found'; })()`);
+    await new Promise(r => setTimeout(r, 50));
+  }
+  return await _nativeTypeViaClipboard(value);
+}
+
+export async function nativeKeyboard({ key, modifiers = [] }) {
+  await ensureHelpers();
+  if (!key) throw new Error("nativeKeyboard requires 'key'");
+  const k = String(key).toLowerCase();
+  const keyCode = macKeyCodeMap[k];
+  if (keyCode === undefined) {
+    throw new Error(`nativeKeyboard: unsupported key "${key}". Supported: ${Object.keys(macKeyCodeMap).join(", ")}`);
+  }
+  const geo = await _getSafariWindowGeometry();
+  if (!geo.windowId) throw new Error("Cannot native-key without Safari window ID — would steal focus");
+  const normalized = (modifiers || []).map(m => String(m).toLowerCase());
+  await _helperNativeKeyboard(keyCode, normalized, geo.windowId);
+  const modsLabel = normalized.length ? normalized.join("+") + "+" : "";
+  return `Native key: ${modsLabel}${k} (CGEvent to window ${geo.windowId}, no focus steal)`;
+}
+
 // System Events key codes — only used for paste_image, upload_file, save_pdf
 // (functions that truly require OS-level UI interaction)
 
@@ -1523,13 +1579,16 @@ export async function pressKey({ key, modifiers = [] }) {
   return `Pressed: ${modifiers.length ? modifiers.join("+") + "+" : ""}${key}`;
 }
 
-// Native typing via clipboard paste — for cross-origin iframes (Intercom, Zendesk, etc.)
-// JS can't access content inside cross-origin iframes, so we use OS-level CGEvent:
-// 1. Save current clipboard
-// 2. Set clipboard to our text
-// 3. Cmd+V via CGEvent keyboard (targeted to window — NO focus steal, NO activate)
-// 4. Restore clipboard
-// Requires nativeClick to have focused the input inside the iframe first.
+// ========== NATIVE TYPE VIA CLIPBOARD PASTE ==========
+// Uses OS-level clipboard + CGEvent Cmd+V to insert text. Produces a REAL paste event
+// that ProseMirror/Slate/Draft.js process through their native paste handlers, updating
+// internal model state (not just the DOM). Also works for cross-origin iframes.
+// NO focus steal — sends CGEvent Cmd+V targeted to Safari window ID.
+//
+// Why this matters: synthetic DOM manipulation (safari_fill's "synthetic paste") writes
+// to the DOM but doesn't update React/ProseMirror state. Discord's onSubmit reads from
+// state, not DOM, so Enter submits empty. This function fixes that by going through the
+// real paste pipeline.
 async function _nativeTypeViaClipboard(text) {
   await _acquireClipboardLock();
   try {