skills.rest
mgsystemsdev
mgsystemsdevCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

swarm

Coordinate multiple AI agents into executable plans with schema v2.

Updated Apr 19, 2026
Search Tags:#workflow#planning#runtime#multi-agent#orchestrator#dag
One-click install
npx skills add https://github.com/mgsystemsdev/POD --skill swarm-mgsystemsdev
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: swarm
Source: https://github.com/mgsystemsdev/POD/tree/main/agent-system-base/.claude/skills/swarm
Command: npx skills add https://github.com/mgsystemsdev/POD --skill swarm-mgsystemsdev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Executable multi-agent plan orchestration that coordinates multiple AI agents as a single, cohesive workflow with shared state and a centralized merge artifact.

Core Features & Use Cases

  • Orchestrates a defined DAG of agents with a single orchestrator and an API runner
  • Supports plan and schema v2, simulated and real execution modes, and merge conflict handling
  • Ideal for coordinating complex AI research workflows across multiple agents in a project

Quick Start

Point the orchestrator at your project root and run in simulate mode to preview the plan.

Frequently Asked Questions about swarm

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

FAQPage Schema
How do I orchestrate a multi-agent workflow with shared state?▼

Multi-agent workflow orchestration coordinates multiple AI agents into a single executable plan with shared state and a centralized merge artifact, using an API runner to enforce the execution logic. It targets project roots with claude-system.

What is the best way to coordinate AI agents across a DAG?▼

Coordinating AI agents across a DAG is handled by a single orchestrator and an API runner that validate agent outputs against a defined schema v2 metadata contract, managing execution and merges.

How do I preview a multi-agent plan before real execution?▼

To preview a multi-agent plan before real execution, point the orchestrator at your project root and run it in simulate mode to preview the plan safely.

Can I handle merge conflicts when coordinating multiple AI agents?▼

Yes, you can handle merge conflicts during multi-agent coordination. The orchestrator supports merge conflict handling alongside simulated and real execution modes across DAGs.

Does multi-agent plan orchestration support handoff workflows?▼

Yes, multi-agent plan orchestration supports handoff workflows alongside simulate and execute modes, allowing seamless transitions between agents within a defined DAG.

What is a metadata contract in multi-agent plan orchestration?▼

A metadata contract in multi-agent plan orchestration is a schema v2 validation mechanism that the API runner enforces to validate agent outputs within the DAG.

Related Skills

Explore More Software Engineering Skills→
WenyuChiouWenyuChiou
23

agent-task-splitter

Split a high-level goal into a DAG of subtasks and emit .coord/plan.yml plus per-agent task files.

Community
Advanced
chaimtchaimt

architect-implement

Generate a unified AD.md from ADRs using multi-agent DAG orchestration.

Community
Advanced
hamzaPixlhamzaPixl
2

multi-agent-pipeline

Coordinate domain-specific LLM agents with an orchestrator and runner for structured outputs.

Community
Advanced
Pear-CommercePear-Commerce

gsd-autonomous

Automates GSD milestone workflow phases from discussion to execution.

Official
Advanced