# Verification as a separate step for computer-use agents: our pattern

**URL:** <https://forum.langchain.com/t/verification-as-a-separate-step-for-computer-use-agents-our-pattern/4874>\
**Category:** Talking Shop\
**Created:** [October 27, 2026, 3:53pm UTC](https://forum.langchain.com/t/verification-as-a-separate-step-for-computer-use-agents-our-pattern/4874 "2026-10-27T15:53:14Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![NovastartOS](https://yyz1.discourse-cdn.com/flex007/user_avatar/forum.langchain.com/novastartos/32/3714_2.png) [@NovastartOS](https://forum.langchain.com/u/NovastartOS)\
**Post date:** [October 27, 2026, 3:53pm UTC](https://forum.langchain.com/t/verification-as-a-separate-step-for-computer-use-agents-our-pattern/4874/1 "2026-10-27T15:53:14Z")

</div>

Sharing a pattern from building Novastart ([https://novastart.com/](https://novastart.com/)), where up to 10 agents work on a desktop canvas, each with its own mouse and keyboard. A lead agent drafts a plan with one task per helper, and humans can pause or type a correction mid-run. The key piece is that \*done means checked\*: a separate check inspects a fresh screenshot before a task is marked complete, independent of the agent’s own report. How are others structuring verification in LangGraph: a separate node, a separate model, or human review?