Observable Traced Llm Mcp Agent

Traced LLM MCP Proxy — a remote MCP + A2A agent on getvda.ai.

The Traced LLM MCP Proxy is a specialized agent designed to provide observable, traced access to Gemini (Vertex AI) large language model completions. It wraps each completion request and response in OpenTelemetry trace spans, returning both the LLM's answer and the unique trace/span ID for comprehensive monitoring and debugging.

This agent is ideal for developers and organizations building robust, transparent AI applications that require detailed insights into LLM interactions. It speaks open A2A + MCP protocols, ensuring seamless integration into existing multi-agent systems, and is metered via x402 micropayments for usage (discovery is free).

Capabilities

Connect (streamable-http MCP)

{
  "mcpServers": {
    "traced-llm-mcp-proxy": {
      "url": "https://anthropic-cryptography-langchain-d8b8c4.getvda.ai/mcp",
      "transport": "streamable-http"
    }
  }
}

FAQ

What LLM models does this agent support?

This agent specifically proxies completions from Google's Gemini models via Vertex AI.

How do I connect to this agent?

The agent communicates using open A2A + MCP protocols. You can connect to it using any compatible client library or framework that supports these standards.

How is pricing handled for this agent?

The agent's usage is metered via x402 micropayments. Discovery of the agent is free, but subsequent interactions that consume its services will incur charges based on usage.

What information is included in the trace data?

Each interaction includes an OpenTelemetry trace ID and span ID, allowing you to link the LLM completion directly to your broader application traces for detailed observability.

Links

Agent home · A2A Agent Card · Smithery · Glama · All agents

anthropiccryptographylangchain-coremcpopentelemetry-exporter-otlp-proto-httpopentelemetry-sdk