prose

Execute .prose programs that orchestrate multi-agent workflows via the Task tool.

6|Updated Mar 13, 2026

Search Tags:#automation #workflow #multi-agent #prose #orchestration #vm

Ready to run

npx skills add https://github.com/me2Doc/friendlyclaw --skill prose-me2doc

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: prose
Source: https://github.com/me2Doc/friendlyclaw/tree/main/body/extensions/open-prose/skills/prose
Command: npx skills add https://github.com/me2Doc/friendlyclaw --skill prose-me2doc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill solves the complexity of managing multi-agent AI workflows by providing a structured, Turing-complete programming language that turns your AI session into a virtual machine.

Core Features & Use Cases

Multi-Agent Orchestration: Define and coordinate multiple specialized agents to complete complex, multi-step tasks.
Persistent Workflows: Maintain state across sessions and projects using file-based or database-backed memory.
Use Case: Build a complex research pipeline where one agent gathers data, a second agent analyzes the findings, and a third agent writes a final report, all managed through a single .prose program.

Quick Start

Use the prose skill to run the example program located at 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 build multi-agent AI workflows that maintain state across sessions?▼

Multi-agent workflows that maintain state across sessions are built by orchestrating subagent sessions via the Task tool using a virtual machine that executes .prose programs, enabling persistent memory and parallel execution.

What is the best way to coordinate multiple specialized agents for complex automation tasks?▼

The best way to coordinate specialized agents for complex automation is using a Turing-complete programming language that turns your AI session into a virtual machine, enabling modular program composition and parallel execution.

How do I run a .prose program to start an automation workflow?▼

To run a .prose program, use the prose skill to execute an example program such as examples/01-hello-world.prose, which requires the OpenProse VM specification to interpret and execute within the AI session.

Do I need the OpenProse VM specification to execute multi-agent programs?▼

Yes, you need the OpenProse VM specification to interpret and execute .prose programs within an AI session, as it provides the structured language required to spawn subagent sessions and manage persistent state.

Can I use file-based memory to maintain persistent workflows across multiple projects?▼

Yes, persistent workflows can be maintained across projects using file-based or database-backed memory, allowing complex multi-step tasks like research pipelines to preserve state between different agent sessions.