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

# Integrate a Play into your application

> Have your agent integrate Play results into your application.

[Agent guide: CLI and SDK instructions →](/docs/sdk-v2/production-integration)

## 1. Have your context ready

Describe the application’s latency needs, result destination, and which external writes you permit.

## 2. Ask your agent

Paste this into your agent, then provide your task and example inputs when it asks:

```deepline-prompt theme={null}
/deepline-gtm Design the Deepline Play integration for my application. Choose whether the caller reads saved results, receives a callback, or the Play writes to an approved destination. Preserve launch intent and run identity across restarts. Explain how timeouts, duplicate delivery, failures, and cancellation will be handled. Show a small test and the proposed external writes before deploying anything.
```

Complete any requested sign-in yourself. Give the agent the missing information before approving execution. Provider-backed work can incur usage.

## 3. Review the result

The application reconciles the saved run and destination state after interruption. It does not assume that a timed-out request means nothing happened.

Ask your agent:

```deepline-prompt theme={null}
/deepline-gtm Compare the result with the acceptance rule we agreed. Show the evidence for accepted and rejected outcomes, unresolved questions, and recorded Deepline usage where work ran. If work is still running or a response is uncertain, inspect the original saved execution before proposing another request. Explain the next decision I need to make.
```

Approve a larger job, recurring work, or external changes only after reviewing that decision.

## Check your understanding

If a callback times out after submission, can you assume the destination made no change?

> **Show the answer**
>
> No. Preserve the uncertain outcome and reconcile it. The destination may have
> accepted the action before the response was lost.

## Related guides

* [Recover a tool execution](/docs/human/sdk-v2/tool-execution-recovery)
* [Call Plays](/docs/human/sdk-v2/calling-plays)
* [Background Agents](/docs/human/sdk-v2/background-agents)
* [HTTP and Python](/docs/human/sdk-v2/http-and-python)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.