# My Real Multi-Agent Workflow in Orca: Smooth & Autonomous Collaboration

**URL:** <https://www.artofsm.art/t/my-real-multi-agent-workflow-in-orca-smooth-autonomous-collaboration/24189>\
**Category:** Content Creators\
**Tags:** hermes, onchain-ai-garage, agents, automation\
**Created:** [22 September 2026 00:00 UTC](https://www.artofsm.art/t/my-real-multi-agent-workflow-in-orca-smooth-autonomous-collaboration/24189 "2026-09-22T00:00:00Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![artesia](https://www.artofsm.art/user_avatar/www.artofsm.art/artesia/32/36_2.png) [@artesia](https://www.artofsm.art/u/artesia)\
**Post date:** [22 September 2026 00:00 UTC](https://www.artofsm.art/t/my-real-multi-agent-workflow-in-orca-smooth-autonomous-collaboration/24189/1 "2026-09-22T00:00:00Z")

</div>

[![](https://www.artofsm.art/uploads/default/original/3X/7/1/71062783b16c4a50a06a355e83a87523c46c48a7.jpeg "My Real Multi-Agent Workflow in Orca: Smooth & Autonomous Collaboration") ](https://www.youtube.com/watch?v=8mlDqA8HLeE)

The video demonstrates a real multi-agent workflow using Orca, where two AI agents, Claude and Hermes, autonomously collaborate on redesigning an iOS educational app by dividing tasks and continuously reviewing each other’s work through Orca’s command line interface. The creator highlights the simplicity, efficiency, and smoothness of this two-agent setup compared to more complex systems, emphasizing its potential to streamline project management with minimal human intervention.

---

<div class="post-metadata">

**Author:** ![artesia](https://www.artofsm.art/user_avatar/www.artofsm.art/artesia/32/36_2.png) [@artesia](https://www.artofsm.art/u/artesia)\
**Post date:** [26 September 2026 06:44 UTC](https://www.artofsm.art/t/my-real-multi-agent-workflow-in-orca-smooth-autonomous-collaboration/24189/3 "2026-09-26T06:44:14Z")

</div>

In this video, the creator demonstrates a real multi-agent workflow using Orca, showcasing a smooth and autonomous collaboration between two AI agents—Claude and Hermes—working on an old mobile learning app project. The creator explains how Hermes manages Claude’s work through Orca’s command line interface (CLI), enabling seamless communication and task division between the two agents. This setup allows the agents to work simultaneously on different aspects of the project, with Claude focusing on planning and design, while Hermes handles coding, testing, and visual asset creation. The creator emphasizes the simplicity and clarity of using just two agents rather than complex multi-agent systems.

The workflow begins with Claude reviewing the existing project and outlining a redesign plan for the iOS educational app, which is inspired by Duolingo but focused on AI and agents. Hermes then analyzes Claude’s plan and suggests corrections, leading to a collaborative agreement on task division. They use Orca CLI commands to communicate, read each other’s outputs, and coordinate their work without much human intervention. This collaboration results in a shared work\_split.md file that clearly defines ownership of different codebase paths, preventing conflicts and ensuring smooth parallel development.

Throughout the project, the agents autonomously review each other’s work, exchange feedback, and update the codebase accordingly. The creator highlights how this process is remarkably smooth compared to other multi-agent environments, with minimal need for manual prompting or corrections. Hermes and Claude effectively monitor each other’s progress and maintain continuous communication, which the creator attributes either to Orca’s design or the complementary strengths of the two AI models. This autonomous interaction significantly reduces the creator’s workload in managing the project.

The creator also discusses the broader orchestration capabilities of Orca, such as task launches, managed executors, and checkpoints, but prefers a simpler setup with just two main agents for clarity and ease of management. They compare Orca with another multi-agent platform, Herder, noting that while Herder excels in session recovery, Orca offers better project-specific workspace management. The creator plans to use both platforms in parallel, leveraging Orca for long-term projects and Herder for daily, smaller tasks.

In conclusion, the video showcases an effective multi-agent workflow in Orca that enables autonomous, harmonious collaboration between two powerful AI agents on a real-world project. The creator expresses enthusiasm about the smoothness and efficiency of this setup and hints at future videos detailing the development of the educational app. They invite viewers to share their experiences with Orca and multi-agent orchestration, emphasizing the potential of such workflows to streamline complex development tasks with minimal human oversight.