zo-swarm-orchestrator

Name: zo-swarm-orchestrator
Availability: InStock
Author: Marlando

Orchestrate multi-agent tasks locally with deterministic routing and memory.

23|4|Updated Mar 27, 2026

Search Tags:#routing #delegation #agents #memory #swarm #orchestrator #dags

Ready to run

npx skills add https://github.com/marlandoj/zouroboros --skill zo-swarm-orchestrator

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: zo-swarm-orchestrator
Source: https://github.com/marlandoj/zouroboros/tree/main/packages/swarm
Command: npx skills add https://github.com/marlandoj/zouroboros --skill zo-swarm-orchestrator

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @agentclientprotocol/sdk, hono, zouroboros-core, and includes scripts (resource) components.

What problem does it solve?

Zouroboros Swarm is a local-first multi-agent orchestrator that coordinates tasks, memory, and execution with deterministic prompts, reducing manual overhead and enabling autonomous task completion.

Core Features & Use Cases

6-Signal Composite Routing for deterministic executor selection across capabilities, health, complexity, history, procedure, and time-based signals.
Hierarchical Orchestration with delegation to decompose complex tasks while preserving isolation and telemetry.
Local bridges to Claude Code, Hermes, Gemini, and Codex for fast, offline execution with memory integration and auto-episodes.
RAG/memory grounding to ground tasks in project context and improve prompt relevance (opt-in).

Quick Start

Run bun Skills/zo-swarm-orchestrator/scripts/orchestrate-v5.ts tasks.json to launch your first swarm campaign.

Frequently Asked Questions about zo-swarm-orchestrator

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

FAQPage Schema

How do I coordinate multi-agent tasks locally with memory and deterministic routing?▼

Multi-agent tasks can be coordinated locally using an orchestrator that applies hierarchical delegation and 6-signal routing to decompose complex workflows while preserving execution isolation and telemetry.

What is the best way to route prompts to specific AI executors in a local swarm?▼

Routing prompts to specific AI executors is best handled using 6-signal composite routing, which evaluates capabilities, health, complexity, history, procedure, and time-based signals for deterministic executor selection.

Does this multi-agent orchestrator support local execution bridges for Claude Code and Gemini?▼

Yes, the orchestrator supports local executor bridges for Claude Code, Hermes, Gemini, and Codex, enabling fast offline execution with memory integration and auto-episodes.

How do I integrate multi-agent orchestration with my AI assistant using MCP transport?▼

You can integrate multi-agent orchestration with an AI assistant by utilizing the built-in MCP transport and local executor bridges, ensuring seamless workflow coordination and context preservation.

When should I use hierarchical delegation for complex multi-agent workflows?▼

Hierarchical delegation should be used to decompose complex tasks into manageable subtasks, optimizing throughput and reliability while maintaining strict isolation and telemetry across the execution swarm.

Can I ground multi-agent tasks in project context using RAG memory?▼

Yes, you can ground tasks in project context using opt-in RAG and memory grounding features, which improve prompt relevance by anchoring execution to historical project data.