---
title: "Comment by Vlad Zoff on \"How Do I Debug an AI Agent Run After It Finished?\" | Hashnode"
description: "The replay idea is useful, but full traces can become another copy of production data. Tool inputs and outputs may contain API tokens, customer records, emails"
url: "https://hashnode.com/posts/how-do-i-debug-an-ai-agent-run-after-it-finished/6aea60ce421b4fe5e077c984/comment/6aec23745c230ec8005922e0"
---

[Discussion](https://hashnode.com/posts/how-do-i-debug-an-ai-agent-run-after-it-finished/6aea60ce421b4fe5e077c984)

Comment [How Do I Debug an AI Agent Run After It Finished?](https://hashnode.com/posts/how-do-i-debug-an-ai-agent-run-after-it-finished/6aea60ce421b4fe5e077c984)

[Vlad Zoff](https://hashnode.com/@vladzoff)

[Nov 5](https://hashnode.com/posts/how-do-i-debug-an-ai-agent-run-after-it-finished/6aea60ce421b4fe5e077c984/comment/6aec23745c230ec8005922e0)

The replay idea is useful, but full traces can become another copy of production data. Tool inputs and outputs may contain API tokens, customer records, emails or other sensitive values.

I'd want field-level redaction before events are persisted, not just a warning to be careful when exporting them. Do you have a way to configure that, or is it currently up to the application to sanitize the payloads?
