publish website

[aojea]

No description provided.

[gemini-code-assist]

Code Review

This pull request migrates the project's documentation from Docsify to a Hugo/Docsy-based site structure under the site/ directory, updating paths, styling, and configuration accordingly. It also introduces a new Python example (gemini_agent.py) demonstrating how to integrate a Google Gemini AI agent with the Sovereign Agent Mesh (SAM) using the Model Context Protocol (MCP). The review feedback highlights critical authentication and concurrency issues: first, the mcp-client commands in the quickstart guide need to include the authentication token to avoid unauthorized errors; second, the synchronous input() call in the asynchronous Gemini agent script should be run in a separate thread to prevent blocking the event loop; and third, the Gemini integration tutorial should instruct users to export the SAM_API_TOKEN environment variable for successful client connection.

[gemini-code-assist]

The local sam-node is configured with an API token (--api-token my-secret-token) by default in the Quick Start guide. To allow the Gemini agent to connect successfully, the SAM_API_TOKEN environment variable must be exported so that SamClient can authenticate with the node.

Suggested change:

3. **Gemini API Key & SAM API Token**: Obtain an API key from Google AI Studio and export both keys:
   ```bash
   export GEMINI_API_KEY="your-api-key-here"
   export SAM_API_TOKEN="my-secret-token"