> ## 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.

# Process a CSV with your agent

> Run a reviewed workflow on a file without losing row identity.

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

## 1. Have your context ready

Have a CSV, the result you need, and a stable source identifier. Start with representative records rather than the full file.

## 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 my CSV and the Deepline workflow that could process it. Parse the file as CSV, including quoted values and embedded newlines. Preserve a stable source ID. Prepare a separate pilot with representative valid, incomplete, and ambiguous records. Explain missing fields, output columns, acceptance rules, and expected usage. Show the pilot before starting provider work.
```

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

Every pilot source ID has a result or an explicit exception. Compare accepted results, misses, errors, and recorded usage before approving the same method for the full file.

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

One quoted address spans two lines in your file. Can the agent select records by counting physical lines?

> **Show the answer**
>
> No. Ask it to parse CSV records and preserve the header and source identities
> in a separate pilot file.

## Related guides

* [Validate Incoming CSV Files Before CRM Enrichment](/docs/human/guides/validate-incoming-csv-files)
* [Write Plays](/docs/human/sdk-v2/plays)


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