native-devtools-mcp 0.3.5 → 0.3.6
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +4 -3
- package/package.json +3 -3
package/README.md
CHANGED
|
@@ -54,7 +54,7 @@ This MCP server is designed to be **highly discoverable and usable** by AI model
|
|
|
54
54
|
**Core Capabilities for System Prompts:**
|
|
55
55
|
1. `take_screenshot`: The "eyes". Returns images + layout metadata + text locations (OCR).
|
|
56
56
|
2. `click` / `type_text`: The "hands". Interacts with the system based on visual feedback.
|
|
57
|
-
3. `find_text`: A shortcut to find text on screen and get its coordinates immediately.
|
|
57
|
+
3. `find_text`: A shortcut to find text on screen and get its coordinates immediately. Uses the platform **accessibility API** (macOS Accessibility / Windows UI Automation) for precise element-level matching, with OCR fallback.
|
|
58
58
|
4. `load_image` / `find_image`: Template matching for non-text UI elements (icons, shapes), returning screen coordinates for clicking.
|
|
59
59
|
|
|
60
60
|
## 📦 Installation (macOS + Windows)
|
|
@@ -202,11 +202,12 @@ graph TD
|
|
|
202
202
|
|----|---------|----------|
|
|
203
203
|
| **macOS** | Screenshots | `screencapture` (CLI) |
|
|
204
204
|
| | Input | `CGEvent` (CoreGraphics) |
|
|
205
|
+
| | Text Search (`find_text`) | `Accessibility API` (primary), Vision OCR (fallback) |
|
|
205
206
|
| | OCR | `VNRecognizeTextRequest` (Vision Framework) |
|
|
206
207
|
| **Windows** | Screenshots | `BitBlt` (GDI) |
|
|
207
208
|
| | Input | `SendInput` (Win32) |
|
|
209
|
+
| | Text Search (`find_text`) | `UI Automation` (primary), WinRT OCR (fallback) |
|
|
208
210
|
| | OCR | `Windows.Media.Ocr` (WinRT) |
|
|
209
|
-
| | Text Search (`find_text`) | `UI Automation` (primary), OCR (fallback) |
|
|
210
211
|
|
|
211
212
|
### Screenshot Coordinate Precision
|
|
212
213
|
|
|
@@ -256,7 +257,7 @@ On macOS, you must grant permissions to the **host application** (e.g., Terminal
|
|
|
256
257
|
|
|
257
258
|
Works out of the box on **Windows 10/11**.
|
|
258
259
|
* Uses standard Win32 APIs (GDI, SendInput).
|
|
259
|
-
* `find_text` uses **UI Automation (UIA)** as the primary search mechanism, querying the accessibility tree for element names. This is
|
|
260
|
+
* `find_text` uses **UI Automation (UIA)** as the primary search mechanism, querying the accessibility tree for element names. This is the same accessibility-first approach used on macOS (with the Accessibility API). Falls back to OCR automatically when UIA finds no matches.
|
|
260
261
|
* OCR uses the built-in Windows Media OCR engine (offline).
|
|
261
262
|
* **Note:** Cannot interact with "Run as Administrator" windows unless the MCP server itself is also running as Administrator.
|
|
262
263
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "native-devtools-mcp",
|
|
3
|
-
"version": "0.3.
|
|
3
|
+
"version": "0.3.6",
|
|
4
4
|
"description": "MCP server for computer-use / desktop automation of native apps (screenshots, OCR, input)",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"repository": {
|
|
@@ -39,8 +39,8 @@
|
|
|
39
39
|
"bin"
|
|
40
40
|
],
|
|
41
41
|
"optionalDependencies": {
|
|
42
|
-
"@sh3ll3x3c/native-devtools-mcp-darwin-arm64": "0.3.
|
|
43
|
-
"@sh3ll3x3c/native-devtools-mcp-win32-x64": "0.3.
|
|
42
|
+
"@sh3ll3x3c/native-devtools-mcp-darwin-arm64": "0.3.6",
|
|
43
|
+
"@sh3ll3x3c/native-devtools-mcp-win32-x64": "0.3.6"
|
|
44
44
|
},
|
|
45
45
|
"engines": {
|
|
46
46
|
"node": ">=18"
|