kanban-orchestrator

Route and decompose multi-profile work into Kanban tasks without executing them.

Updated Jun 1, 2026
One-click install
npx skills add https://github.com/SatangThevalue/ai-skills --skill kanban-orchestrator-satangthevalue
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kanban-orchestrator
Source: https://github.com/SatangThevalue/ai-skills/tree/main/skills/devops/kanban-orchestrator
Command: npx skills add https://github.com/SatangThevalue/ai-skills --skill kanban-orchestrator-satangthevalue

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Decomposition playbooks and anti-temptation rules empower an orchestrator to route work through Kanban without performing tasks themselves, ensuring scalable, crash-resistant task distribution across profiles.

Core Features & Use Cases

  • Decomposition playbook: breaks user requests into independent lanes and maps them to capable profiles.
  • Anti-temptation rules: prevents execution by the orchestrator, preserving specialist ownership.
  • Robust guidance: auto-injects core lifecycle patterns into each Kanban worker to ensure consistent behavior.
  • Use cases include multi-profile routing, parallel task fan-out, and governance for long-running tasks.

Quick Start

Ground the decomposition by discovering available profiles (e.g., hermes profile list) and then create Kanban tasks for each lane, letting the dispatcher handle task distribution.

Frequently Asked Questions about kanban-orchestrator

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

FAQPage Schema
How do I orchestrate multi-agent workflows without letting the orchestrator execute tasks itself?

Multi-agent workflow orchestration uses anti-temptation rules to prevent the orchestrator from executing tasks, preserving specialist ownership. A decomposition playbook breaks user requests into independent lanes mapped to capable profiles for compliant routing.

What is the best way to decompose long-running tasks to survive system restarts?

Decomposing long-running tasks is handled via kanban_create and parent task linking to model dependencies. This robust guidance auto-injects core lifecycle patterns into each Kanban worker, ensuring crash-resistant task distribution across profiles.

How do I set up parallel task fan-out across multiple specialist profiles?

Parallel task fan-out is set up by discovering available profiles, creating Kanban tasks for each decomposed lane, and letting the dispatcher handle task distribution. This maps independent work lanes directly to capable specialist profiles.

Can I route work through Kanban when human-in-the-loop approval is required?

Kanban routing supports human-in-the-loop requirements by modeling dependencies through parent task linking. The orchestrator distributes work across profiles without executing tasks, ensuring specialists retain ownership of approval steps.

Why does my orchestrator keep executing tasks instead of routing them to specialists?

Orchestrators execute tasks when anti-temptation rules are absent or bypassed. Implementing a decomposition playbook enforces strict task routing via kanban_create, preventing the orchestrator from performing specialist work itself.