# Traced Llm Proxy

> Traced Llm Proxy is an MCP server for ai models and llm tooling. LLM proxy with OpenTelemetry tracing, exposed over MCP and A2A. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Platforms:** API, Web
- **Website:** https://traced-llm-proxy.getvda.ai
- **MCP Registry name:** ai.getvda/traced-llm-proxy
- **Categories:** [AI models and LLM tooling](https://ai-powered.com/categories/ai-and-llm), [developer tools](https://ai-powered.com/categories/developer-tools)
- **Listing updated:** 2026-11-18

## About

Traced LLM Proxy is a remote service from getvda.ai that combines the Anthropic client library, the MCP SDK and OpenTelemetry API and SDK packages. Its site describes it as covering LLM client access, agent protocol support and observability.

It is reachable through a streamable-HTTP MCP endpoint and also publishes an Agent Card for the Agent2Agent (A2A) protocol. The site also links to a governance page and notes a listing on Smithery, and it belongs to a wider portfolio of agents at getvda.ai.

The page gives no list of individual tools, so the exact operations it exposes are not documented. It is best suited to developers who want LLM calls routed through a service that adds tracing.

## Features

- **Remote MCP endpoint:** Connects to MCP clients over streamable HTTP at a hosted URL, with no local package to install.
- **A2A Agent Card:** Publishes an Agent Card so other agents can discover it via the Agent2Agent protocol.
- **OpenTelemetry observability:** Built on the OpenTelemetry API and SDK to provide tracing of LLM activity.
- **Anthropic client integration:** Uses the Anthropic library as its LLM client layer.
- **Governance page:** Links to a governance section alongside the agent and MCP endpoints.

## Good for

- Tracing LLM calls made by an MCP-enabled agent
- Adding observability to Anthropic-based workflows
- Letting A2A agents discover and call a shared LLM proxy
- Experimenting with composed agent-protocol services

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http traced-llm-proxy https://traced-llm-proxy.getvda.ai/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://traced-llm-proxy.getvda.ai/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "traced-llm-proxy": {
      "url": "https://traced-llm-proxy.getvda.ai/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "traced-llm-proxy": {
      "type": "http",
      "url": "https://traced-llm-proxy.getvda.ai/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "traced-llm-proxy": {
      "serverUrl": "https://traced-llm-proxy.getvda.ai/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "traced-llm-proxy": {
      "type": "http",
      "url": "https://traced-llm-proxy.getvda.ai/mcp"
    }
  }
}
```

## FAQ

### How do I connect to Traced LLM Proxy?

Use the streamable-HTTP MCP endpoint at https://traced-llm-proxy.getvda.ai/mcp.

### Does it support the A2A protocol?

Yes, the site links to an A2A Agent Card in addition to the MCP endpoint.

### What libraries is it built from?

It composes anthropic, mcp, opentelemetry-api and opentelemetry-sdk.

### Where else is it listed?

The site notes a listing on Smithery and links to a portfolio at getvda.ai/agents.

### How do I add Traced Llm Proxy to Claude Code?

Run this command in your terminal: claude mcp add --transport http traced-llm-proxy https://traced-llm-proxy.getvda.ai/mcp

### What are the best alternatives to Traced Llm Proxy?

The closest alternatives to Traced Llm Proxy are Traced Llm MCP Proxy, Traced Multi Llm Proxy, Llm Orchestration Traced Agent, Llm Orchestration Agent MCP and Traced Openai Langchain Agent.

## Where this comes from

- Description: written from the project's website and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [Traced Llm MCP Proxy](https://ai-powered.com/mcp/traced-llm-mcp-proxy): Traced Llm MCP Proxy is an MCP server for ai models and llm tooling.
- [Traced Multi Llm Proxy](https://ai-powered.com/mcp/traced-multi-llm-proxy): Traced Multi Llm Proxy is an MCP server for ai models and llm tooling.
- [Llm Orchestration Traced Agent](https://ai-powered.com/mcp/llm-orchestration-traced-agent): Llm Orchestration Traced Agent is an MCP server for ai models and llm tooling.
- [Llm Orchestration Agent MCP](https://ai-powered.com/mcp/llm-orchestration-agent-mcp): Llm Orchestration Agent MCP is an MCP server for ai models and llm tooling.
- [Traced Openai Langchain Agent](https://ai-powered.com/mcp/traced-openai-langchain-agent): Traced Openai Langchain Agent is an MCP server for ai models and llm tooling.
- [Llm Orchestration Agent 2 Df5979](https://ai-powered.com/mcp/llm-orchestration-agent-2-df5979): Llm Orchestration Agent 2 Df5979 is an MCP server for ai models and llm tooling.
- [Structured Output MCP Traced Agent](https://ai-powered.com/mcp/structured-output-mcp-traced-agent): Structured Output MCP Traced Agent is an MCP server for developer tools.
- [Llm Orchestration Agent D0a1a1](https://ai-powered.com/mcp/llm-orchestration-agent-d0a1a1): Llm Orchestration Agent D0a1a1 is an MCP server for ai models and llm tooling.

---
Source: https://ai-powered.com/mcp/traced-llm-proxy
