> ## 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.

# Find work emails with your agent

> Ask your agent to find work emails and stop only at results that meet your rule.

[Agent guide: CLI and SDK instructions →](/docs/plays/email-waterfall)

## 1. Have your context ready

Have a small CSV with first names, last names, company domains, and stable source IDs. Decide which verification outcomes you will accept.

## 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 Inspect a maintained Deepline name-and-company work-email waterfall for this file. Check the required columns and validation behavior. Create a separate representative pilot using a proper CSV parser. Preserve source IDs, source evidence, validation outcomes, and explicit no-result reasons. Show me the route and expected usage before running it. Do not send emails or upload the output to a sequencer.
```

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

Accepted emails meet your verification policy and match the intended person and company. Catch-all, unknown, rejected, and provider-error outcomes remain distinct.

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

Does a completed batch imply every row contains an accepted work email?

> **Show the answer**
>
> No. Inspect per-row outcomes. A legitimate no-result or rejection can remain
> within a completed execution.

## Related guides

* [Example Prebuilt Plays](/docs/human/plays/tasks-overview)
* [Cost control](/docs/human/cost-control)
* [Plays](/docs/plays/plays-overview)


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