Deploy a2a-starter-agent

Minimal, LLM-agnostic A2A protocol agent for any A2A client

Deploy a2a-starter-agent

Just deployed

Deploy and Host A2A Starter Agent on Railway

A minimal A2A protocol agent, backed by an LLM of your choice. Deploy it, paste the resulting URL into any A2A client (e.g. AgentChat Hub), and start chatting immediately.

Before You Deploy: Choose Your LLM Provider

Set these variables on the deploy screen based on the provider you want:

ProviderLLM_PROVIDERBASE_URLMODEL (example)Get an API key
Anthropic (Claude)anthropic(leave blank)claude-sonnet-4-5console.anthropic.com
OpenAIcustomhttps://api.openai.com/v1gpt-4oplatform.openai.com
xAI (Grok)customhttps://api.x.ai/v1grok-4console.x.ai
Moonshot (Kimi)customhttps://api.moonshot.ai/v1kimi-k2platform.moonshot.ai
DeepSeekcustomhttps://api.deepseek.com/v1deepseek-chatplatform.deepseek.com
Ollama / LM Studio (self-hosted)customyour server's URL, e.g. http://your-host:11434/v1whatever model you have pullednone needed — leave API_KEY blank
Any other OpenAI-compatible APIcustomthat provider's base URLthat provider's model idthat provider's dashboard

API_KEY is always required except for an unauthenticated local server. Everything else has a sensible default and can be left as-is.

About Hosting A2A Starter Agent

This template runs a small FastAPI server exposing the two endpoints an A2A client needs: GET /.well-known/agent.json (the agent card) and POST /a2a (JSON-RPC message/send). Incoming chat messages are forwarded to the LLM you configure above, and the reply is returned in A2A's message format. There's no database, no queue, and no other services — just one web process.

Why Deploy A2A Starter Agent on Railway?

Railway lets you go from "I want an agent" to a live, publicly-reachable A2A endpoint in one deploy, with your own LLM API key and no infrastructure to manage — a public domain, HTTPS, and restarts are all handled for you.

Common Use Cases

  • A ready-to-chat starting point for anyone building or testing an A2A client
  • A base to fork and extend with your own tools, prompts, or model
  • A quick way to give any LLM-backed agent a public A2A endpoint

Dependencies for A2A Starter Agent Hosting

Deployment Dependencies

  • An API key from your chosen LLM provider (see the table above) — required, set as the API_KEY variable

Template Content

More templates in this category

View Template
Chat Chat
Chat Chat, your own unified chat and search to AI platform.

okisdev
116
View Template
stella
Self-host stella with web, API, Postgres, Redis, and object storage.

Jan Kubica
5
View Template
Hermes Agent | OpenClaw Alternative with Dashboard
Self-Hosted Hermes AI Agent for Telegram, Discord & Slack

codestorm
67