WCMD

v0.6.0 suspicious
6.0
Medium Risk

WinControl MCP Driver — 讓 AI Agent 用自然語言操控 Windows UI 的 MCP 伺服器

🤖 AI Analysis

Final verdict: SUSPICIOUS

The package exhibits signs of obfuscation and has metadata indicating recent creation and low activity, raising concerns about its legitimacy and potential for misuse.

  • High obfuscation risk due to base64 decoding and JPEG header checks
  • Metadata suggests a newly created and less active repository
Per-check LLM notes
  • Network: No network calls detected, which is normal and does not indicate any risk.
  • Shell: Detection of shell execution may be benign if the package involves running scripts, but it could also be a potential risk for executing arbitrary code. Further investigation is needed.
  • Obfuscation: The code shows multiple instances of base64 decoding and checking for JPEG headers, which may indicate obfuscation or hiding of data.
  • Credentials: No clear patterns indicative of credential harvesting were found in the provided code snippets.
  • Metadata: The package is suspicious due to the recent creation of both the repository and the package, along with the low activity and a new maintainer account.

🔬 Heuristic Checks

✓ Outbound Network Calls

No suspicious network call patterns found

⚠ Code Obfuscation score 10.0

Found 5 obfuscation pattern(s)

  • 解碼回 JPEG bytes 確認是 JPEG raw = base64.b64decode(b64) expect_true(raw[:2] == b"\xff\xd8", "解碼後前 2
  • small = Image.open(io.BytesIO(base64.b64decode(b64_small))) expect(img_back_small.width, 1280, "1280 寬不縮放")
  • t_realistic_4k.png")) raw_v = base64.b64decode(b64_v) expect_true(raw_v[:2] == b"\xff\xd8", "EC-1a: 輸出是 JPE
  • img_s = Image.open(io.BytesIO(base64.b64decode(b64_s))) expect(img_s.width, 1280, "EC-1e: 1280 寬不縮放") expec
  • ambda rows=10, cols=10: ( __import__("PIL").Image.new("RGB", (1000, 800), "white"), {f"{engine.GRID
⚠ Shell / Subprocess Execution score 2.0

Found 1 shell execution pattern(s)

  • YTHONPATH", "") result = subprocess.run( [sys.executable, str(script_path)], capture
✓ Credential Harvesting

No credential harvesting patterns detected

✓ Typosquatting

No typosquatting candidates detected

✓ Registered Email Domain

No author email provided

✓ Suspicious Page Links

All external links appear legitimate

⚠ Git Repository History score 7.5

Git history flags: Repository created very recently: 1 day(s) ago (2026-06-04T17:21:14Z)

  • Repository created very recently: 1 day(s) ago (2026-06-04T17:21:14Z)
  • Repository has zero stars and zero forks
  • All 22 commits happened within 24 hours
⚠ Maintainer History score 4.0

2 maintainer concern(s) found

  • Package is very new: uploaded 1 day(s) ago
  • Author "leowu0511" appears to have only 1 package on PyPI (new or inactive account)
✓ Known CVE Vulnerabilities

No known vulnerabilities found in OSV database.

💡 AI App Starter Prompt

Use this prompt to build a project with WCMD 
Your task is to develop a simple yet powerful desktop automation tool using the Python package 'WCMD'. This tool will allow users to control their Windows User Interface through natural language commands, making it easier to automate repetitive tasks and improve productivity. Here’s a detailed breakdown of your project:

1. **Project Name:** Natural Language Desktop Controller (NLDC)
2. **Objective:** Build a Python application that leverages WCMD to interpret user commands and execute corresponding actions on the Windows desktop.
3. **Core Features:**
   - **Command Interpretation:** Implement a basic command parser that understands natural language inputs like 'open Microsoft Word', 'close Chrome', 'type hello world', etc.
   - **Action Execution:** Use WCMD to translate these parsed commands into actions that manipulate the Windows UI, such as opening applications, typing text into fields, clicking buttons, etc.
   - **User Interaction:** Provide a simple GUI or CLI interface for users to input commands and receive feedback about action execution.
4. **Implementation Steps:**
   - **Step 1:** Set up your development environment with Python and install the WCMD package.
   - **Step 2:** Design the structure of your command parser. Decide on a set of predefined commands and how they map to specific actions.
   - **Step 3:** Integrate WCMD into your application. Learn how to use its functions to interact with the Windows UI.
   - **Step 4:** Develop a user-friendly interface for inputting commands and displaying results.
   - **Step 5:** Test your application thoroughly with various commands to ensure reliability and accuracy.
5. **Suggested Enhancements:**
   - Add support for more complex commands, such as navigating menus or filling out forms.
   - Incorporate machine learning models to improve command recognition over time.
   - Allow for scripting of multiple commands to create automated workflows.
6. **Deliverables:**
   - A fully functional NLDC application.
   - Documentation detailing the setup process, command syntax, and any additional features implemented.
   - A report summarizing the challenges faced during development and potential improvements.

This project will not only showcase your ability to work with cutting-edge technologies but also contribute to making computer interactions more intuitive and efficient.