prose

Author, validate, run, and debug OpenProse programs with a deterministic VM runtime.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/RJB07/RX01 --skill prose-rjb07
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prose
Source: https://github.com/RJB07/RX01/tree/main/core/extensions/open-prose/skills/prose
Command: npx skills add https://github.com/RJB07/RX01 --skill prose-rjb07

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

OpenProse provides a structured runtime for orchestrating AI agents and workflows, replacing ad-hoc prompts with a formal VM-driven approach that manages state, parallelism, and dependencies across tasks.

Core Features & Use Cases

  • Robust multi-agent orchestration with a single, uniform VM model.
  • Supports sessions, imports, parallel execution, loops, error handling, and persistent agent memories.
  • Use cases include building orchestrated QA pipelines, multi-agent research, and automated content workflows across projects.

Quick Start

To begin, load prose.md in this directory and run a sample program with prose run examples/01-hello-world.prose.

Frequently Asked Questions about prose

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

FAQPage Schema
How do I orchestrate AI agents with deterministic state management?

AI agent orchestration uses a deterministic VM-like runtime to execute .prose programs, replacing ad-hoc prompts with formal state management. It handles parallel execution, loops, and persistent memories to automate multi-agent workflows.

What is the best way to run parallel execution loops in multi-agent workflows?

Parallel execution loops in multi-agent workflows run through a deterministic VM-like runtime that natively supports loops and concurrent tasks. This manages dependencies and state across concurrent agents without ad-hoc prompting.

Does OpenProse support persistent memory backends like sqlite or postgres?

Yes, OpenProse supports persistent memory backends including sqlite and postgres, alongside filesystem and in-context options. These backends allow orchestrated agents to maintain state and context across sessions.

Can I use a single VM model for automated content workflows and QA pipelines?

A single uniform VM model supports automated content workflows and QA pipelines through sessions and imports. The runtime orchestrates diverse automation scenarios by managing state and dependencies across tasks.

How do I debug runtime state and validate .prose programs?

Debug runtime state and validate .prose programs using a guided reference that enables VM runtime introspection. The runtime allows you to author, run, and debug OpenProse programs while inspecting execution state.

Why use a VM-driven approach over ad-hoc prompts for workflow orchestration?

A VM-driven approach provides structured orchestration by managing state, parallelism, and dependencies formally, unlike ad-hoc prompts. This ensures deterministic execution and robust automation across complex multi-agent tasks.