# Railway for AI agents. Managed auth, ready-made tools.

> List projects and services, read deployment logs, redeploy and set variables in Railway on behalf of each of your users.

- Page: https://www.agent200.com/connectors/railway/
- Category: Developer tools
- Authentication: Managed OAuth
- Identifier in code: `railway`
- Create an API key: https://agent200.com/platform/apikeys/
- Updated: 2026-11-18

## Instructions for AI agents

Connector: Railway, through Agent200. In code: "railway".
What it does: List projects and services, read deployment logs, redeploy and set variables in Railway on behalf of each of your users.
Page: https://www.agent200.com/connectors/railway/

1. Get an Agent200 API key. If the developer has none, send them to https://agent200.com/platform/apikeys/ and ask them to choose "Create an API key". Read the key from the AGENT200_API_KEY environment variable. Never ask for it in chat, and never write it into code, prompts or logs.
2. Sign-in is managed OAuth. Each end user approves access once, on Railway's own page, through the authorization link Agent200 returns. Never ask a user for a Railway password or token in chat.
3. Give every request a stable id for the end user. Agent200 uses only that user's own connection.
4. Read tools: railway_list_projects, railway_list_services, railway_list_deployments, railway_get_deployment_logs. Write tools: railway_redeploy_deployment, railway_restart_deployment, railway_upsert_variable. Name "railway" in the tools of a request; the model decides which tool to call.
5. Stay within the project's settings. If a tool is turned off or a call is refused, tell the user. Do not work around it.
6. Before a write, confirm with the user, unless they asked for that exact change.
7. For setup with a specific framework, follow its guide:
   https://www.agent200.com/connectors/railway/framework/claude/
   https://www.agent200.com/connectors/railway/framework/chatgpt/
   https://www.agent200.com/connectors/railway/framework/claude-code/
   https://www.agent200.com/connectors/railway/framework/muse/
   https://www.agent200.com/connectors/railway/framework/grok-bot/
   https://www.agent200.com/connectors/railway/framework/openai-dots/
   https://www.agent200.com/connectors/railway/framework/claude-duet/
   https://www.agent200.com/connectors/railway/framework/hermes-agent/
   https://www.agent200.com/connectors/railway/framework/openclaw/

The code on the connector page is a conceptual example of the SDK. SDK reference: https://docs.igpt.ai/

## About

With the Railway connector, your agent works in each user's own Railway account. It can list projects and services, check recent deployments and their logs, redeploy a service and set an environment variable, using only the tools your project allows.

Railway is a cloud platform for deploying apps, databases and services from code.

## Railway tools

- **List projects** (`railway_list_projects`, read): List the user's Railway projects.
- **List services** (`railway_list_services`, read): List the services in a project.
- **List deployments** (`railway_list_deployments`, read): List recent deployments of a service with their status.
- **Get deployment logs** (`railway_get_deployment_logs`, read): Read the build or runtime logs of a deployment.
- **Redeploy** (`railway_redeploy_deployment`, write): Redeploy an existing deployment.
- **Restart a deployment** (`railway_restart_deployment`, write): Restart a running deployment.
- **Set a variable** (`railway_upsert_variable`, write): Create or update an environment variable for a service.

## Limits you can set

**Turn single tools on or off.** Let the agent read deployments and logs, and switch off `railway_upsert_variable` so it can never change an environment variable.

**Restrict what the agent reaches.** Choose the endpoints and scopes a project may use, for example deployments and logs but not variables. A call outside them is refused.

## Get started

### Step 1: Set up the SDK

Add the Agent200 SDK to the codebase you already have and authenticate with your API key. The key covers every connector enabled for the project.

File `agent.js`:

```js
// One Agent200 API key covers every connector enabled for the project.
const agent200 = new Agent200({
  apiKey: process.env.AGENT200_API_KEY,
});
```

(Conceptual example of the Agent200 SDK.)

### Step 2: Let your user connect Railway

Each end user authorizes their own Railway account once. Agent200 runs the OAuth flow, handles the callback, stores the tokens and refreshes them.

File `connect.js`:

```js
// Each end user authorizes their own Railway account, once.
const connection = await agent200
  .user("user_123")
  .connect("railway");

if (connection.authorizationUrl) {
  // Show this link to your user; Agent200 handles the rest.
  console.log(connection.authorizationUrl);
}
```

(Conceptual example of the Agent200 SDK.)

### Step 3: Run with the Railway tools

Name the user, the model and the tools. The model decides which Railway tools to call, and Agent200 executes them with that user's connection.

File `run.js`:

```js
const result = await agent200
  .user("user_123")
  .run({
    model: "openai/gpt-5.6-sol",
    input: "Why did the last deploy of my API fail",
    tools: ["railway"],
  });

console.log(result.output);
```

(Conceptual example of the Agent200 SDK.)

## Sign-in and permissions

**Who signs in.** Each of your users, once, on Railway's own sign-in page. You never see or store their Railway password.

**What access is requested.** The scopes the connection needs, such as reading projects, reading deployments or managing services, listed on Railway's authorization screen. Users can remove the app later from their Railway account settings.

## Example requests

- "Why did the last deploy of my API fail"
- "Show the logs for the worker service"
- "Restart the web service in production"
- "Set LOG_LEVEL to debug on the staging API"

## Guides: Railway with your AI framework

- [Railway + Claude](https://www.agent200.com/connectors/railway/framework/claude.md): Set up Railway for Claude with Agent200, then check which tools are available to your agent.
- [Railway + ChatGPT](https://www.agent200.com/connectors/railway/framework/chatgpt.md): Set up Railway for ChatGPT with Agent200, then check which tools are available to your agent.
- [Railway + Claude Code](https://www.agent200.com/connectors/railway/framework/claude-code.md): Set up Railway for Claude Code with Agent200, then check which tools are available to your agent.
- [Railway + Muse](https://www.agent200.com/connectors/railway/framework/muse.md): Set up Railway for Muse with Agent200, then check which tools are available to your agent.
- [Railway + Grok Bot](https://www.agent200.com/connectors/railway/framework/grok-bot.md): Set up Railway for Grok Bot with Agent200, then check which tools are available to your agent.
- [Railway + OpenAI Dots](https://www.agent200.com/connectors/railway/framework/openai-dots.md): Set up Railway for OpenAI Dots with Agent200, then check which tools are available to your agent.
- [Railway + Claude Duet](https://www.agent200.com/connectors/railway/framework/claude-duet.md): Set up Railway for Claude Duet with Agent200, then check which tools are available to your agent.
- [Railway + Hermes Agent](https://www.agent200.com/connectors/railway/framework/hermes-agent.md): Set up Railway for Hermes Agent with Agent200, then check which tools are available to your agent.
- [Railway + OpenClaw](https://www.agent200.com/connectors/railway/framework/openclaw.md): Set up Railway for OpenClaw with Agent200, then check which tools are available to your agent.

## FAQ

### Who handles the Railway OAuth flow and the tokens?

Agent200. It runs the authorization, handles the callback on its own domain, stores the tokens and refreshes them. You do not build or maintain an OAuth integration for Railway.

### Does my agent act as the end user?

Within what that user authorized, yes. Every request names an end user, and Agent200 uses only that user's Railway connection, never another user's. It cannot reach anything the user or Railway has not authorized.

### Can I limit what my agent can do in Railway?

Yes. You choose which services, endpoints and authorization scopes are allowed in your Agent200 configuration, and the agent works only within them.

## Related connectors

- [GitHub](https://www.agent200.com/connectors/github.md): Read repositories and work with issues and pull requests as the connected user.
- [GitLab](https://www.agent200.com/connectors/gitlab.md): Review merge requests, file issues, read code and check pipelines in your users' GitLab projects.
- [Bitbucket](https://www.agent200.com/connectors/bitbucket.md): List and open pull requests, comment, read code and check pipelines in your users' Bitbucket workspaces.
- [Jira](https://www.agent200.com/connectors/jira.md): Search, create and transition Jira issues for the connected user.

---

Agent200: https://www.agent200.com/ · API keys: https://agent200.com/platform/apikeys/ · Docs: https://docs.igpt.ai/
