open-prose

Define and execute AI agent workflows using the OpenProse programming language.

6|1|Updated Sep 22, 2025

Search Tags:#workflow automation #ai agents #orchestration #control flow #openprose #programming language

Ready to run

npx skills add https://github.com/enyst/OpenHands-Tab --skill open-prose-enyst

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: open-prose
Source: https://github.com/enyst/OpenHands-Tab/tree/main/.openhands/skills/open-prose
Command: npx skills add https://github.com/enyst/OpenHands-Tab --skill open-prose-enyst

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides a programming language (OpenProse) to structure complex AI workflows, turning natural language prompts into unambiguous control flow for AI agents.

Core Features & Use Cases

Declarative Workflow Definition: Define multi-agent tasks using a clear, English-like syntax.
Agent Orchestration: Structure sequences, parallel tasks, loops, and error handling for AI agents.
Use Case: Automate a multi-stage code review process involving security, performance, and style analysis agents, then synthesize their feedback into a single report.

Quick Start

Use the open-prose skill to run the example program 'examples/01-hello-world.prose'.

Frequently Asked Questions about open-prose

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

FAQPage Schema

How do I structure multi-agent AI workflows using natural language?▼

To structure multi-agent AI workflows using natural language, use the OpenProse programming language to turn English prompts into unambiguous control flow for agent orchestration. It enables declarative definition of complex multi-stage tasks.

Can I run parallel tasks and loops in AI agent orchestration?▼

Yes, you can run parallel tasks and loops in AI agent orchestration. OpenProse explicitly manages control flow including sequential execution, parallel tasks, loops, and error handling to coordinate multiple specialized agents.

What is the best way to automate a multi-stage code review process with AI agents?▼

The best way to automate a multi-stage code review process is by defining specialized security, performance, and style analysis agents in OpenProse, then orchestrating them to synthesize feedback into a single report.

Do I need prior programming knowledge to define AI workflow control flow?▼

You need some programming knowledge to define AI workflow control flow, as OpenProse is a programming language utilizing concepts like loops, error handling, and reusable blocks, but it uses an English-like syntax for declarative definitions.

How do I execute an example AI workflow program to test agent orchestration?▼

To execute an example AI workflow program and test agent orchestration, use the open-prose skill to run the provided example file 'examples/01-hello-world.prose' to see the declarative workflow mechanism in action.

How does context passing work between specialized AI agents in a workflow?▼

Context passing between specialized AI agents works by managing the state and data flow through the OpenProse control flow structure. This ensures unambiguous data handoffs during sequential, parallel, and looped agent execution.