prose

Execute structured multi-agent workflows using declarative prose syntax.

4|2|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/Clawdi-AI/openclaw --skill prose-clawdi-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prose
Source: https://github.com/Clawdi-AI/openclaw/tree/main/extensions/open-prose/skills/prose
Command: npx skills add https://github.com/Clawdi-AI/openclaw --skill prose-clawdi-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill allows you to define and execute complex multi-agent workflows using a structured, yet natural language-like programming paradigm, turning AI sessions into powerful, programmable computers.

Core Features & Use Cases

  • Agent Orchestration: Define custom agents with specific models and prompts, then coordinate them.
  • Workflow Automation: Build sequential, parallel, or conditional execution flows for AI tasks.
  • State Management: Persist execution state across sessions using files or databases for complex, long-running processes.
  • Use Case: Automate a code review process where one agent identifies issues, another suggests fixes, and a third synthesizes the feedback into a report.

Quick Start

Use the prose skill to run the example program '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 for complex multi-agent workflows?

You orchestrate AI agents by defining and executing structured programs using a declarative syntax. The system manages agent definitions, parallel execution, loops, and error handling to coordinate complex multi-agent workflows.

Can I persist state across sessions for long-running AI agent workflows?

Yes, you can persist execution state across sessions using files or databases. This state management capability enables complex, long-running multi-agent processes to maintain data continuity between executions.

How do I run parallel execution and conditional logic in LLM orchestration?

Parallel execution and conditional logic are handled natively by defining structured programs with a declarative syntax. You coordinate sequential, parallel, or conditional execution flows to automate complex AI tasks.

What's the best way to automate a multi-step code review process with AI agents?

You can automate a code review process by defining custom agents with specific models and prompts, then coordinating them. One agent identifies issues, another suggests fixes, and a third synthesizes feedback into a report.

Does managing AI agents require a specific programming language or framework?

No external dependencies are required to manage AI agents. The system uses a declarative syntax within structured programs to define custom agents and execute recursive language model patterns without external frameworks.

When should I not use declarative syntax for AI agent orchestration?

Declarative syntax for AI agent orchestration is not ideal for simple, single-prompt tasks that lack multi-step logic. It is designed for complex workflows requiring session spawning, loops, and state persistence.