From 4d5114741faa25cbfab1e72e0af46835f3ae0111 Mon Sep 17 00:00:00 2001
From: twzrd <33047129+twzrd-sol@users.noreply.github.com>
Date: Sat, 6 Jun 2026 16:23:01 -0500
Subject: [PATCH] samples(mcp): add TWZRD Agent Intel trust verification
 concept sample

---
 .../agent_with_twzrd_trust_verification.py    | 84 +++++++++++++++++++
 1 file changed, 84 insertions(+)
 create mode 100644 python/samples/concepts/mcp/agent_with_twzrd_trust_verification.py

diff --git a/python/samples/concepts/mcp/agent_with_twzrd_trust_verification.py b/python/samples/concepts/mcp/agent_with_twzrd_trust_verification.py
new file mode 100644
index 000000000000..0564b7533ba8
--- /dev/null
+++ b/python/samples/concepts/mcp/agent_with_twzrd_trust_verification.py
@@ -0,0 +1,84 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import asyncio
+
+from semantic_kernel.agents import ChatCompletionAgent, ChatHistoryAgentThread
+from semantic_kernel.connectors.ai.open_ai import OpenAIChatCompletion
+from semantic_kernel.connectors.mcp import MCPStreamableHttpPlugin
+
+"""
+This sample demonstrates how to create a Semantic Kernel agent with access
+to the TWZRD Agent Intel MCP server (https://intel.twzrd.xyz) for trust scoring
+and x402 payment verification of AI agents on Solana.
+
+TWZRD Agent Intel is a zero-install remote MCP server that provides:
+- score_agent(wallet): Returns trust score (0-100) and reputation data
+- resolve_agent(wallet): Agent identity resolution
+- preflight_check(wallet): Pre-transaction safety check
+- verify_trust_receipt(receipt): Verify x402 payment receipts
+
+MCP config: {"mcpServers": {"twzrd-agent-intel": {"url": "https://intel.twzrd.xyz/mcp"}}}
+
+Required environment variables:
+- OPENAI_API_KEY
+- OPENAI_CHAT_MODEL_ID (e.g. gpt-4o-mini)
+"""
+
+# Example agent wallets to check
+USER_INPUTS = [
+    "What is the trust score for Solana wallet D1QkbFJKiPsymJ65RKHhF6DFB8sPMfpBaFBzuHKfJGWi?",
+    "Run a preflight check on wallet 4Nd1mFxXgSU4LJSS2aWUKf6vXhiGaWMPBJXQWgpjvPdT before proceeding.",
+]
+
+
+async def main():
+    # 1. Create TWZRD Agent Intel MCP plugin (no API key needed — free tier)
+    async with MCPStreamableHttpPlugin(
+        name="TwzrdAgentIntel",
+        description="Trust scoring and x402 payment verification for AI agents on Solana",
+        url="https://intel.twzrd.xyz/mcp",
+    ) as twzrd_plugin:
+        # 2. Create the agent with TWZRD tools available
+        agent = ChatCompletionAgent(
+            service=OpenAIChatCompletion(),
+            name="TrustVerificationAgent",
+            instructions=(
+                "You are a trust verification agent. "
+                "Use the TWZRD Agent Intel tools to score agent wallets, run preflight checks, "
+                "and verify x402 payment receipts. "
+                "A trust score >= 70 is high trust, 40-70 is medium, < 40 is low. "
+                "Always recommend caution for wallets with low trust scores."
+            ),
+            plugins=[twzrd_plugin],
+        )
+
+        for user_input in USER_INPUTS:
+            thread: ChatHistoryAgentThread | None = None
+
+            print(f"# User: {user_input}")
+            response = await agent.get_response(messages=user_input, thread=thread)
+            print(f"# {response.name}: {response}\n")
+            thread = response.thread
+
+            await thread.delete() if thread else None
+
+
+"""
+Sample output:
+
+# User: What is the trust score for Solana wallet D1QkbFJKiPsymJ65RKHhF6DFB8sPMfpBaFBzuHKfJGWi?
+# TrustVerificationAgent: I checked the TWZRD Agent Intel trust score for wallet
+  D1QkbFJKiPsymJ65RKHhF6DFB8sPMfpBaFBzuHKfJGWi.
+  Trust Score: 82/100 — High Trust
+  This agent has a strong track record of successful x402 transactions and consistent activity.
+  It is generally safe to interact with this agent.
+
+# User: Run a preflight check on wallet 4Nd1mFxXgSU4LJSS2aWUKf6vXhiGaWMPBJXQWgpjvPdT ...
+# TrustVerificationAgent: Preflight check complete for 4Nd1mFxXgSU4LJSS2aWUKf6vXhiGaWMPBJXQWgpjvPdT.
+  Result: safe_to_transact = false
+  Reason: Wallet has insufficient transaction history for reliable trust scoring.
+  Recommendation: Proceed with caution or request additional identity verification.
+"""
+
+if __name__ == "__main__":
+    asyncio.run(main())