yaml-composer

Build multi-agent orxhestra trees from declarative YAML configuration.

19|6|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/NicolaiLassen/orxhestra --skill yaml-composer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: yaml-composer
Source: https://github.com/NicolaiLassen/orxhestra/tree/main/docs/skills/yaml-composer
Command: npx skills add https://github.com/NicolaiLassen/orxhestra --skill yaml-composer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

yaml-composer removes the hassle of manually wiring complex multi-agent orxhestra configurations by letting you define an entire agent tree, models, and tools declaratively in a single YAML file.

Core Features & Use Cases

  • Declarative agent trees: Build full multi-agent setups (models, tools, agents, and main agent) from YAML configuration.
  • Model and tool composition: Define named models and connect tools including builtins, function-based tools, MCP tools, and transfer targets.
  • Runner and server configuration: Configure runner and FastAPI server settings, including exposing skills via server YAML.

Quick Start

Ask yaml-composer to generate a working orxhestra multi-agent configuration from your orx.yaml file and then run it to stream responses in Python.

Frequently Asked Questions about yaml-composer

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

FAQPage Schema
How do I generate a multi-agent setup from a YAML configuration file?

You can generate a multi-agent setup from YAML by defining your entire agent tree, models, and tools declaratively in a single YAML file, which the parser converts into a runnable composer-based workflow. This removes the hassle of manual wiring for complex agent graphs.

What is declarative agent tree composition for multi-agent orchestration?

Declarative agent tree composition is the process of defining models, selecting agent types, and attaching tools in a YAML configuration file rather than writing procedural code. It enables building complete multi-agent setups by parsing the YAML into structured agent graph objects.

Can I configure MCP tools and transfer targets in a YAML agent graph?

Yes, you can configure MCP tools and transfer targets within your YAML agent graph. The declarative configuration supports connecting builtins, function-based tools, MCP tools, and transfer targets directly alongside your named model definitions.

How do I configure a FastAPI server for interactive multi-agent workflows?

You configure a FastAPI server by defining runner and server settings within your YAML configuration file. This setup allows you to expose skills via server YAML and supports both interactive and API-driven execution for your multi-agent workflows.

Do I need a specific schema to parse YAML into multi-agent composer objects?

Yes, you need a SKILL.md-defined YAML composer schema to parse YAML into composer objects correctly. This schema provides the functional support required for accurate agent graph construction and workflow generation.

What are the limitations of building multi-agent setups with YAML configuration?

A limitation of building multi-agent setups with YAML configuration is the strict dependency on a SKILL.md-defined composer schema for correct agent graph construction. Without this schema, the parser cannot accurately turn declarative agent files into runnable workflows.