Authenticated Llm Mcp Agent

Authenticated MCP Agent — a remote MCP + A2A agent on getvda.ai.

This authenticated MCP agent provides a secure gateway to powerful Large Language Models. It specifically leverages LangChain to orchestrate Gemini completions on Google Cloud's Vertex AI, but only after rigorous authentication via JWTs. Unauthenticated requests are automatically rejected, ensuring controlled and secure access to your AI resources.

Designed for developers and organizations requiring enterprise-grade security for their AI integrations, this agent speaks open A2A (Application to Application) and MCP (Metering and Control Protocol). All interactions are metered via x402 micropayments, with discovery of the agent itself remaining free.

Capabilities

Connect (streamable-http MCP)

{
  "mcpServers": {
    "authenticated-mcp-agent": {
      "url": "https://bcrypt-langchain-core-mcp-25b8f1.getvda.ai/mcp",
      "transport": "streamable-http"
    }
  }
}

FAQ

How does authentication work for this LLM gateway?

The agent uses JWTs (JSON Web Tokens) for authentication. Calls to the LLM gateway must include a valid JWT in the request headers; otherwise, they are rejected.

What LLM is supported and how is it orchestrated?

This agent specifically supports Google's Gemini models via Vertex AI, orchestrated using LangChain. This allows for complex prompt chaining and interaction patterns.

How are API calls priced or metered?

All API calls through this agent are metered using x402 micropayments. Discovery of the agent itself is free, but subsequent LLM completions will incur charges via x402.

How do I connect to this authenticated LLM MCP agent?

You connect to the agent via its A2A endpoint. You'll need to obtain a valid JWT to include with your requests, following standard A2A and MCP protocols for interaction.

Links

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

bcryptlangchain-coremcpopenaiopentelemetry-sdk