minion-orchestrator

Orchestrate durable background jobs with a Postgres-backed queue.

Updated Jun 2, 2026
One-click install
npx skills add https://github.com/Ninatuzi/gbrain --skill minion-orchestrator-ninatuzi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: minion-orchestrator
Source: https://github.com/Ninatuzi/gbrain/tree/main/skills/minion-orchestrator
Command: npx skills add https://github.com/Ninatuzi/gbrain --skill minion-orchestrator-ninatuzi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of unreliable background task execution by providing a Postgres-native, durable queue that ensures jobs survive system restarts and remain observable throughout their lifecycle.

Core Features & Use Cases

  • Unified Job Management: Handles both deterministic shell scripts and complex LLM subagent reasoning loops through a single interface.
  • Steerable Execution: Allows users to pause, resume, cancel, or send mid-flight directives to running agents to change their behavior dynamically.
  • Use Case: Use this to spawn a research subagent that performs parallel web searches and synthesizes a report, while maintaining the ability to steer the agent if it drifts off-topic or to pause it to save tokens.

Quick Start

Submit a new background subagent job to research a specific topic by running the gbrain agent run command followed by your prompt.

Frequently Asked Questions about minion-orchestrator

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

FAQPage Schema
How do I orchestrate durable background jobs using Postgres?

Orchestrate LLM subagent reasoning loops by submitting prompts that execute deterministic shell commands and iterative reasoning tasks in the background. Jobs run through a Postgres-backed queue, enabling parallel fan-out and structured progress tracking.

Can I pause or steer a running background subagent mid-flight?

Yes, you can steer running subagents mid-flight by sending directives to pause, resume, or cancel active jobs. This allows you to dynamically change agent behavior to prevent off-topic drift or conserve tokens during execution.

Do I need a configured worker environment to run background orchestration jobs?

Yes, background orchestration requires a configured worker environment with appropriate tool allow-lists and persistent database connectivity. You must establish this setup to manage job states and execute tasks safely across sessions.

What is the best way to run parallel subagent research tasks in the background?

You can handle both deterministic shell scripts and complex LLM subagent reasoning loops through a single unified job management interface. This consolidates background task execution and lifecycle monitoring into one system.

Why use a Postgres-backed queue for background job orchestration?

Use Postgres-backed queues for background job orchestration when you need reliable task execution that survives system restarts. Avoid it if your environment lacks persistent database connectivity or configured worker tool allow-lists.