kanban-orchestrator

Decompose goals into specialist Kanban tasks with dependency-gated promotion.

Updated May 26, 2026
One-click install
npx skills add https://github.com/ruiyangruiyi/hermes-agent --skill kanban-orchestrator-ruiyangruiyi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kanban-orchestrator
Source: https://github.com/ruiyangruiyi/hermes-agent/tree/main/skills/devops/kanban-orchestrator
Command: npx skills add https://github.com/ruiyangruiyi/hermes-agent --skill kanban-orchestrator-ruiyangruiyi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps you orchestrate complex work by decomposing goals into specialist tasks on a Kanban board, preventing you from falling into the trap of doing everything yourself.

Core Features & Use Cases

  • Decomposition into a task graph: Break ambiguous or multi-part goals into a structured set of dependent tasks (fan-out, fan-in, and gated promotion).
  • Specialist roster routing conventions: Assign work to the right profile (researcher, analyst, writer, reviewer, backend/frontend/ops, pm) using consistent conventions.
  • Anti-temptation guardrails: Enforce the “route, don’t execute” behavior by creating and assigning Kanban tasks for any concrete task.
  • Operational reliability patterns: Use tenant inheritance and recovery workflows (reclaim, reassign, change profile model) for stuck or failing workers.

Quick Start

Tell the orchestrator to route your goal through Kanban by asking it to decompose the work into parallel specialist tasks and show you the proposed task graph before creating any board entries.

Frequently Asked Questions about kanban-orchestrator

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

FAQPage Schema
How do I decompose complex multi-step goals into specialist tasks on a Kanban board?

To decompose complex goals, route work through Kanban by breaking ambiguous objectives into a structured task graph with fan-out, fan-in, and dependency-gated promotion, assigning each item to a specialist profile rather than executing directly.

How can I prevent my AI orchestrator from executing tasks itself instead of delegating?

To prevent self-execution, apply anti-temptation guardrails that enforce a route-don't-execute behavior, ensuring the orchestrator creates and assigns concrete Kanban tasks for any work item rather than attempting to perform the work directly.

What is the best way to recover stalled or crashing workers in a multi-agent task workflow?

Recover stalled workers using operational reliability patterns like tenant inheritance and recovery workflows, which allow you to reclaim, reassign, or change the profile model for failing tasks to restore workflow progress.

How do I set up human-in-the-loop review gates within an automated task workflow?

Set up human-in-the-loop checkpoints by applying dependency-gated promotion to your task graph, ensuring work pauses for explicit review gates before dependent downstream tasks are allowed to begin processing.

Does multi-agent routing require explicit specialist profiles for task assignment?

Yes, multi-agent routing uses specialist roster routing conventions to assign work to the right profile, such as researcher, analyst, writer, reviewer, backend, frontend, ops, or pm, ensuring tasks align with the correct specialist capabilities.

What are the limitations of relying on an orchestrator for task decomposition?

The orchestrator is limited to coordination only; it cannot execute concrete tasks itself and relies entirely on routing conventions to assign work to specialist profiles, meaning it requires available workers to process the decomposed task graph.