A2A Observable Traced Llm Service

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

The Traced LLM Proxy is a production-ready remote agent that provides access to Gemini (Vertex AI) large language model completions. It automatically wraps each completion request in OpenTelemetry trace spans, returning the LLM's response alongside the unique trace and span identifiers, enabling deep observability into your LLM interactions.

Designed for developers, this service speaks open A2A and MCP protocols, ensuring seamless integration into existing microservices architectures. Metered via x402 micropayments, its discovery is completely free, allowing you to explore its capabilities without upfront costs.

Capabilities

Connect (streamable-http MCP)

{
  "mcpServers": {
    "traced-llm-proxy": {
      "url": "https://anthropic-mcp-opentelemetry-api-264025.getvda.ai/mcp",
      "transport": "streamable-http"
    }
  }
}

FAQ

What LLM models does this service support?

The Traced LLM Proxy specifically supports Gemini completions via Google's Vertex AI.

How do I connect to this A2A observable traced LLM service?

You connect to the Traced LLM Proxy using standard A2A and MCP protocols. Specific connection details and SDKs will be provided upon discovery.

How is this service priced or metered?

This service is metered via x402 micropayments, meaning you pay per use. Discovery of the service itself is free.

What observability data is provided?

For each LLM completion, the service returns the answer from Gemini along with the OpenTelemetry trace ID and span ID, allowing you to correlate LLM activity with your broader system traces.

Links

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

anthropicmcpopentelemetry-apiopentelemetry-sdk