Observable Traced Llm Mcp Agent

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

The Traced LLM Proxy is a specialized agent designed to provide observable traced LLM MCP agent capabilities. It seamlessly proxies Gemini (Vertex AI) completions, enriching them with OpenTelemetry trace spans and returning both the LLM response and the associated trace/span IDs.

This agent speaks open A2A + MCP protocols, ensuring broad compatibility and easy integration. It's ideal for developers and organizations needing detailed observability into their LLM interactions, with metering handled via x402 micropayments for discovery-free access.

Capabilities

Connect (streamable-http MCP)

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

FAQ

What is the pricing model for this agent?

The Traced LLM Proxy agent is metered via x402 micropayments. This allows for discovery-free access, where you pay per use based on the agent's operations.

How do I connect to this observable traced LLM MCP agent?

You connect to this agent using standard A2A (Agent-to-Agent) and MCP (Micro-Agent Communication Protocol) protocols. Specific connection details will be provided upon agent discovery.

Can I use this agent with other LLMs besides Gemini (Vertex AI)?

This specific Traced LLM Proxy agent is designed to proxy Gemini (Vertex AI) completions. Its current composition does not support other LLMs directly without modification.

What information is included in the trace/span ID response?

Upon receiving a completion, the agent returns the LLM's response along with the unique OpenTelemetry trace ID and the specific span ID generated for that particular LLM call, enabling end-to-end observability.

Links

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

anthropicmcpopentelemetry-apiopentelemetry-sdk