> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agentmuxer.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Run in production

> Manage credentials, spending, connection lifecycle, and failures in deployed agents.

Before deploying, check your runtime, credentials, spending settings, and error handling.

## Deployment checklist

<Steps>
  <Step title="Choose a supported runtime">
    Use ESM on Node.js 22.18 or later. See [compatibility](/reference/compatibility) for verified Node and framework versions. The Claude Agent SDK also needs its supported process environment.
  </Step>

  <Step title="Store credentials on the server">
    Set `AGENTMUXER_API_KEY` in your deployment's secret store, or pass the key through `agentmuxer({ apiKey })`. Keep your model-provider key separate. Never send an application key to a browser, mobile bundle, or model prompt.
  </Step>

  <Step title="Set spending and approval policy">
    Applications default to no spending limit and can use their owner's available balance. Set a lifetime application cap when you need a hard cumulative limit. Configure payment confirmation for an interactive or unattended workflow. See [billing and approvals](/guides/billing).
  </Step>

  <Step title="Record useful run information">
    Use your agent framework's tracing to connect a user request to model and tool activity. Retain request and invocation identifiers from AgentMuxer responses for investigation. Keep API keys and authorization headers out of logs and traces.
  </Step>
</Steps>

## Connection lifecycle

| Integration      | Lifecycle                                                                                                        |
| ---------------- | ---------------------------------------------------------------------------------------------------------------- |
| OpenAI Agents    | The helper creates hosted MCP configuration; OpenAI connects to AgentMuxer. There is no SDK connection to close. |
| Claude Agent SDK | The helper creates MCP server configuration. Claude's runner manages the connection.                             |
| Vercel AI SDK    | Await `agentmuxer()`, finish the run or consume the complete stream, then await `close()` in `finally`.          |

Keep each Vercel connection alive for the full run that uses it. If you share a connection, closing it while another run is active interrupts that run. For HTTP streaming, cleanup must run after completion, failure, or cancellation of the stream.

## Timeouts, cancellations, and retries

Set request and run limits using your agent framework. A stopped run or dropped connection doesn't mean a purchase failed. AgentMuxer may already have accepted it.

<Warning>
  Do not retry a purchase with a new idempotency key after a timeout. Preserve the original key when replaying the same purchase request. A new key represents a new purchase.
</Warning>

The Vercel helper disables automatic MCP tool-call retries. When retrying a paid call, keep the original arguments and idempotency key. If you're unsure whether it completed, save the invocation ID and [check the existing purchase](/reference/troubleshooting).

## Handle tool responses

Keep cost, invocation ID, and any refusal or confirmation details when processing tool results.

| Response                               | Your application's next step                                                          |
| -------------------------------------- | ------------------------------------------------------------------------------------- |
| Successful tool result                 | Show or process the result; retain cost and invocation details.                       |
| Authentication failure                 | Check the application key and whether the application or key is disabled/revoked.     |
| Insufficient balance or spending limit | Direct the account owner to the dashboard; do not keep retrying.                      |
| Approval required                      | Resume through a supported confirmation flow, or explain the required account action. |
| Revised price                          | Ask for explicit acceptance of the new terms; do not silently raise the budget.       |
| Uncertain outcome after a timeout      | Keep the original idempotency key and investigate the existing purchase.              |

## Logging and support

The SDK works with your framework's logging and tracing; you do not need an AgentMuxer-specific tracing setup. Capture the SDK and framework versions, your own run identifier, and request/invocation identifiers returned by tools. Avoid logging entire connection options because they can contain your key.

For help, send a minimal reproduction and those identifiers to [founders@amorphiclabs.com](mailto:founders@amorphiclabs.com). Remove secrets and sensitive tool inputs first.

<CardGroup cols={2}>
  <Card title="Applications and keys" icon="key" href="/sdk/applications">
    Rotate credentials and manage application access.
  </Card>

  <Card title="Compatibility and upgrades" icon="arrows-rotate" href="/reference/compatibility">
    Check supported versions before an upgrade.
  </Card>
</CardGroup>