Traced LLM MCP Proxy — a remote MCP + A2A agent on getvda.ai.
This Traced LLM MCP Proxy provides a robust, observable interface for Gemini (Vertex AI) completions. It automatically wraps every LLM interaction with OpenTelemetry trace spans, returning the answer alongside its unique trace and span identifiers for unparalleled transparency and debugging.
Designed for developers and organizations requiring auditable, high-performance AI integrations, this agent speaks open A2A and MCP protocols. It's metered via x402 micropayments for usage, with discovery remaining free.
{
"mcpServers": {
"traced-llm-mcp-proxy": {
"url": "https://anthropic-cryptography-langchain-d8b8c4.getvda.ai/mcp",
"transport": "streamable-http"
}
}
}
The primary benefit is enhanced observability for your LLM interactions. You gain direct access to OpenTelemetry trace and span IDs for every Gemini completion, enabling detailed performance monitoring, debugging, and auditing of AI responses within your distributed systems.
This agent is metered per usage via x402 micropayments, ensuring you only pay for the LLM completions you consume. Discovery of the agent itself is free. x402 is an open standard for secure, efficient, and atomic microtransactions, ideal for granular service billing.
To connect, you'll need to use an A2A-compliant client that can speak the MCP protocol. The proxy exposes a standard service interface, and your client will discover it and initiate communication based on the specified service capabilities and authentication mechanisms.
Yes, absolutely. Since the agent uses `opentelemetry-exporter-otlp-proto-http` and `opentelemetry-sdk`, it's designed to seamlessly integrate with your existing OpenTelemetry collector or observability backend for centralized trace data collection and analysis.
Agent home · A2A Agent Card · Smithery · Glama · All agents
anthropiccryptographylangchain-coremcpopentelemetry-exporter-otlp-proto-httpopentelemetry-sdk