kanban-orchestrator

Route complex multi-step work into dependency-aware Kanban task graphs.

Updated May 14, 2026
One-click install
npx skills add https://github.com/SethyPagna/Secretary-Jarvis --skill kanban-orchestrator-sethypagna
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kanban-orchestrator
Source: https://github.com/SethyPagna/Secretary-Jarvis/tree/main/src/capabilities/skills/devops/kanban-orchestrator
Command: npx skills add https://github.com/SethyPagna/Secretary-Jarvis --skill kanban-orchestrator-sethypagna

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps an AI act as a Kanban orchestrator instead of an executor, so complex requests are decomposed into correctly routed tasks rather than handled all at once.

Core Features & Use Cases

  • Profile Discovery: Identifies which worker profiles actually exist before assigning work.
  • Task Graph Planning: Splits a request into parallel lanes and dependent follow-up cards.
  • Dependency Safety: Uses parent-child links so synthesis, review, and integration steps do not start too early.
  • Use Case: A user asks for research, implementation, and review; this Skill maps the lanes, creates the cards, and keeps the work moving in the right order.

Quick Start

Ask the assistant to break your multi-step request into Kanban tasks by listing the available profiles, mapping each lane to a real assignee, and creating any needed dependencies.

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 requests into Kanban tasks instead of executing them directly?

Multi-step requests are decomposed into Kanban tasks by routing the work into parallel lanes and creating dependent follow-up cards, preventing the AI from executing complex work all at once. This ensures work moves in the correct order.

How does dependency-aware task routing prevent synthesis and review steps from starting too early?

Dependency-aware task routing uses strict parent-child task linking to map task graphs, ensuring synthesis, review, and integration steps do not start until their upstream dependencies are fully resolved. This maintains safe execution order.

Do I need available worker profiles before assigning work to Kanban lanes?

Yes, profile discovery is required to identify which worker profiles actually exist before assigning work. Mapping each lane to a real assignee ensures that routed Kanban tasks are allocated to valid, available workers.

What is the best way to handle fan-out and fan-in coordination for multi-agent task graphs?

Fan-out and fan-in coordination is handled by splitting requests into parallel lanes and creating dependent follow-up cards with strict parent-child links. This orchestrates the task graph so integration steps wait for parallel work to finish.

Can I use Kanban orchestration for recovery-safe task completion without losing work progress?

Kanban orchestration supports recovery-safe completion behavior, allowing complex work routed into dependent Kanban lanes to resume correctly if interrupted. Parent-child task linking ensures the task graph state is preserved.

When should I use a Kanban orchestrator instead of an executor for AI task planning?

Use a Kanban orchestrator instead of an executor when a request requires planning, decomposition, and multi-step coordination. It maps lanes and creates dependency-aware cards so the work moves in the right order rather than executing all at once.