minion-orchestrator

Coordinate durable Postgres-backed job queues for shell tasks and LLM subagents.

Updated May 16, 2026
One-click install
npx skills add https://github.com/Morris-UTRUST/GBrain-knowledge --skill minion-orchestrator-morris-utrust
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: minion-orchestrator
Source: https://github.com/Morris-UTRUST/GBrain-knowledge/tree/main/skills/minion-orchestrator
Command: npx skills add https://github.com/Morris-UTRUST/GBrain-knowledge --skill minion-orchestrator-morris-utrust

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Minions coordinates a durable, observable background job queue for deterministic shell tasks and LLM-driven subagents, enabling reliable orchestration across sessions and restarts.

Core Features & Use Cases

  • Durable, Postgres-native queue with clear lifecycle: submit, progress, steer, pause, resume, cancel, and replay.
  • Support for two lanes: deterministic shell jobs and LLM subagent workflows, with routing between them.
  • Parallelization and fan-out for complex workloads, with structured transcripts and progress reporting.

Quick Start

Submit a shell job using gbrain jobs submit shell to run a deterministic command, or start a subagent workflow with gbrain agent run to orchestrate AI-assisted tasks.

Frequently Asked Questions about minion-orchestrator

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

FAQPage Schema
How do I orchestrate LLM subagents and shell jobs in a background queue?▼

You can orchestrate LLM subagents and shell jobs by submitting them to a durable, Postgres-native background queue that supports parallel fan-out, lifecycle controls, and routing between deterministic shell tasks and AI workflows.

What is the best way to track and steer background subagent workflows across sessions?▼

Tracking and steering background subagent workflows across sessions requires a durable, observable job queue that maintains state through restarts, providing structured transcripts and progress reporting for active workflows.

Can I pause, resume, and cancel deterministic shell tasks in a job queue?▼

Yes, you can pause, resume, cancel, and replay deterministic shell tasks using the queue's lifecycle controls, which manage job states through MCP or CLI endpoints for reliable background execution.

Do I need a Postgres database to manage a durable job queue for LLM subagents?▼

Yes, a Postgres database is required to manage the durable job queue, providing the persistent storage necessary to maintain job state, enable routing rules, and support parallel fan-out across sessions.

How do I submit a shell job or start a subagent workflow using a job queue?▼

You submit a shell job using the CLI command to run a deterministic command, or start a subagent workflow with the agent run command to orchestrate AI-assisted tasks through the queue's endpoints.