prose

Orchestrate multi-agent AI workflows through a VM-like scripting surface.

Updated Apr 24, 2026
One-click install
npx skills add https://github.com/frankhli843/gemmahermes --skill prose-frankhli843
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prose
Source: https://github.com/frankhli843/gemmahermes/tree/main/extensions/open-prose/skills/prose
Command: npx skills add https://github.com/frankhli843/gemmahermes --skill prose-frankhli843

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

OpenProse provides a unified framework to orchestrate multi-agent AI workflows, enabling a single command surface to spawn, coordinate, and manage the execution of subagents across backends and runtimes.

Core Features & Use Cases

  • OpenProse VM surface: Define agents, sessions, parallel blocks, loops, blocks, and imports in a single, self-describing language to orchestrate complex workflows.
  • Stateful orchestration: Tracks execution state, context passing, and memory for persistent agents across runs with filesystem or database backends.
  • Extensible planning patterns: Includes support for blocks, pipelines, error handling, and multi-stage orchestration, enabling production-grade automations.
  • Use Case: Build an orchestrated data-processing pipeline that runs multiple agent reviews in parallel and synthesizes a final report.

Quick Start

Create a simple, end-to-end OpenProse workflow by defining a small agent and a session, then run a session to observe binding creation and context propagation.

Frequently Asked Questions about prose

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I orchestrate multi-agent AI workflows with parallel execution and persistent memory?

You can orchestrate multi-agent AI workflows by using a VM-like scripting surface to define agents, sessions, parallel execution blocks, and memory management across local backends.

What is a structured VM for coordinating AI agents and how does it work?

A structured VM for AI agents provides a self-describing language to define blocks, pipelines, and loops, enabling stateful orchestration and context passing across multiple subagents.

Can I manage stateful orchestration and execution context using filesystem or database backends?

Yes, stateful orchestration tracks execution state and context passing using recommended state backends like filesystem, SQLite, or PostgreSQL to support persistent agents across runs.

How do I build a data-processing pipeline that runs multiple agent reviews in parallel?

To build a parallel data-processing pipeline, define a small agent and a session using the VM surface, then execute the session to observe binding creation, context propagation, and final report synthesis.

Does this multi-agent orchestration framework support error handling and multi-stage automations?

Yes, the framework includes extensible planning patterns that support blocks, pipelines, error handling, and multi-stage orchestration to build robust, reusable, production-grade automations.