orchestrator

Coordinate CLI subagents in parallel with shared state via Serena Memory.

218|33|Updated Jan 7, 2026

Search Tags:#workflow #verification #multi-agent #parallel-execution #orchestrate #serena-memory #cli-automation

Ready to run

npx skills add https://github.com/first-fluke/fullstack-starter --skill orchestrator-first-fluke

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: orchestrator
Source: https://github.com/first-fluke/fullstack-starter/tree/main/.agent/skills/orchestrator
Command: npx skills add https://github.com/first-fluke/fullstack-starter --skill orchestrator-first-fluke

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Automated multi-agent orchestration that spawns and coordinates CLI subagents in parallel, using Serena Memory as a shared state bus to monitor progress.

Core Features & Use Cases

Parallel subagent orchestration: spawn, manage, and limit concurrency with MAX_PARALLEL, MAX_RETRIES, and polling.
Memory-backed coordination: Serena Memory for session-wide state and task board updates.
Lifecycle phases and verification: Plan, Setup, Execute, Monitor, Verify, and Collect with automatic failure handling and result aggregation.
Cross-domain applicability: backend, frontend, mobile, and QA tasks across a full-stack environment.

Quick Start

Start a new orchestration session by executing the orchestrator with a configured task set. Example: place a YAML tasks file listing agents and tasks, then run the orchestrator to spawn and monitor parallel subagents using Serena Memory.

Frequently Asked Questions about orchestrator

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

FAQPage Schema

How do I orchestrate multi-agent tasks in parallel?▼

To orchestrate multi-agent tasks in parallel, use an automation tool to spawn CLI subagents, enforce concurrency limits, and maintain shared state via Serena Memory. Define a YAML task set listing agents and tasks, then execute the orchestrator to manage the lifecycle.

How does Serena Memory coordinate state across parallel subagents?▼

Serena Memory coordinates state across parallel subagents by acting as a shared state bus for session-wide updates. It tracks the task board, enabling automated monitoring of progress, status polling, and result collection across backend, frontend, mobile, and QA domains.

What is the lifecycle for automated multi-agent task execution?▼

The lifecycle for automated multi-agent task execution includes Plan, Setup, Execute, Monitor, Verify, and Collect phases. This sequence handles deterministic task spawning, parallel limits, retries, automatic failure handling, and result aggregation to deliver traceable outcomes.

Can I limit concurrency and retries when spawning CLI subagents?▼

Yes, you can limit concurrency and retries when spawning CLI subagents. The orchestration process enforces deterministic task spawning using configurable MAX_PARALLEL limits and MAX_RETRIES, ensuring controlled parallel execution and automatic failure handling.

Does multi-agent orchestration work for full-stack backend and frontend tasks?▼

Yes, multi-agent orchestration supports full-stack backend and frontend tasks. It provides cross-domain applicability across backend, frontend, mobile, and QA environments, coordinating CLI subagents to run tasks in parallel while maintaining shared state.