Summary

Adds a concept sample demonstrating how to use TWZRD Agent Intel as a remote MCP plugin in a Semantic Kernel agent for trust scoring and x402 payment verification of AI agents on Solana.

File: python/samples/concepts/mcp/agent_with_twzrd_trust_verification.py

What it shows

Uses MCPStreamableHttpPlugin (same pattern as the existing agent_with_http_mcp_plugin.py) to connect to https://intel.twzrd.xyz/mcp and provide a ChatCompletionAgent with TWZRD trust tools:

• score_agent(wallet) — returns 0-100 trust score + reputation data
• preflight_check(wallet) — pre-transaction safety check
• verify_trust_receipt(receipt) — verify x402 payment receipts

Why this fits here

The concepts/mcp/ directory already has agent_with_http_mcp_plugin.py using MCPStreamableHttpPlugin for the Microsoft Learn Docs MCP. TWZRD is another zero-install remote MCP server demonstrating the same connection pattern but for a different domain (Web3/agent trust infrastructure).

Running it

export OPENAI_API_KEY=sk-...
export OPENAI_CHAT_MODEL_ID=gpt-4o-mini
python python/samples/concepts/mcp/agent_with_twzrd_trust_verification.py

No TWZRD API key needed — trust scoring is free.

TWZRD Agent Intel

• MCP endpoint: https://intel.twzrd.xyz/mcp (Streamable HTTP)
• PyPI: pip install twzrd-agent-intel
• MCP config: {"mcpServers": {"twzrd-agent-intel": {"url": "https://intel.twzrd.xyz/mcp"}}}