Gptnix Toolkit

---

⚡ Key Capabilities

• 🧠 Real-Time Grounded Web Research: Autonomously executes multi-step web queries with Google Grounding, returning verified facts and cited source URLs without hallucination.
• 🎨 Multi-Prompt Image Generation: Generates parallel high-fidelity images directly from conversational prompts with built-in creative expansion.
• 👁️ Multimodal Visual QA: Inspects local screenshots, UI mockups, error stack traces, and design diagrams using multimodal vision reasoning.
• 🛠️ Local Zero-Cost Image Utilities: Fast image format conversion (PNG to .ICO, .WEBP), aspect resizing, and compression constraints directly on the host machine without external API billing.
• 📖 Autonomous Codebase Documentation: Ingests large source code contexts into Gemini's high-capacity context window to generate API references and architectural documentation.

---

🧰 Tools Reference

Tool Name	Parameters	Description
web_search	query: str	Live internet search with automatic grounding and cited sources.
generate_image	prompt: str, count: int	Generates high-resolution images matching textual specifications.
vision_analyze	image_path: str, prompt: str	Visual reasoning over local screenshots, diagrams, and logs.
convert_image	source: str, format: str	Host-level format conversion, optimization, and compression.
generate_docs	code_context: str, type: str	Automatic Markdown documentation generation from source code.

---

🏗️ Installation & Setup

Prerequisites

• Python 3.10 or higher
• Valid API key set in AKASH_API_KEY (or compatible endpoint)

# Clone the repository
git clone https://github.com/0xOpCode/gptnix-toolkit.git
cd gptnix-toolkit

# Create virtual environment
python3 -m venv venv
source venv/bin/activate

# Install dependencies
pip install mcp httpx pillow

---

🔌 Client Configuration

Claude Code (~/.claude.json)

{
  "mcpServers": {
    "gptnix-toolkit": {
      "command": "/absolute/path/to/gptnix-toolkit/venv/bin/python",
      "args": [
        "/absolute/path/to/gptnix-toolkit/server.py"
      ],
      "env": {
        "AKASH_API_KEY": "your_api_key_here"
      }
    }
  }
}

Cursor / Windsurf / Cline (mcp.json)

{
  "mcpServers": {
    "gptnix-toolkit": {
      "command": "python",
      "args": ["/absolute/path/to/gptnix-toolkit/server.py"],
      "env": {
        "AKASH_API_KEY": "your_api_key_here"
      }
    }
  }
}

---

⚖️ Architecture & Endpoint Note

This MCP server is pre-configured to communicate via standard Model Context Protocol JSON-RPC endpoints. For Google Grounding integration, ensure the target gateway routes through providers with enabled tools: [{"googleSearch": {}}] to maintain live citation integrity.

---

👨‍💻 Maintainer

Developed by 0xOpCode. Contributions and issues are welcome!