kanban-worker

Guide Hermes Kanban workers through task handoff, lifecycle decisions, and structured reporting.

Updated May 24, 2026
One-click install
npx skills add https://github.com/anxiety135790/hermes-skills --skill kanban-worker-anxiety135790
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kanban-worker
Source: https://github.com/anxiety135790/hermes-skills/tree/main/devops/kanban-worker
Command: npx skills add https://github.com/anxiety135790/hermes-skills --skill kanban-worker-anxiety135790

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps Hermes Kanban workers handle task handoffs, workspace differences, retries, and completion/blocking decisions without losing context or making avoidable workflow mistakes.

Core Features & Use Cases

  • Worker lifecycle guidance: Explains how to orient, work, heartbeat, block, and complete in the Kanban flow.
  • Workspace and tenant handling: Covers scratch, shared directory, and worktree behavior, plus tenant-safe memory prefixing.
  • Handoff quality patterns: Shows strong summary and metadata shapes for coding, research, and review tasks, including when to block for human review.
  • Failure and retry handling: Helps workers interpret blocked, archived, timed out, crashed, or spawn-failed runs and respond appropriately.
  • Use case: A spawned worker needs to decide whether to complete a docs-only task, block a code change for review, or recover from a previous failed run.

Quick Start

Use this skill to understand how to report progress, choose between completion and blocking, and format a clean Kanban handoff for the current worker task.

Frequently Asked Questions about kanban-worker

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

FAQPage Schema
How do I handle task handoffs in a multi-agent Kanban workflow without losing context?

Task handoffs in a multi-agent Kanban workflow require structured summary formatting and metadata capture to maintain context. Workers must follow lifecycle guidance to orient, work, heartbeat, block, or complete tasks while applying tenant-safe memory prefixing.

What is the best way to format worker handoffs for coding and research tasks?

Formatting worker handoffs for coding and research tasks involves using specific summary and metadata shapes. This ensures clean handoff quality patterns that capture progress, workspace differences, and blocking reasons for human review when necessary.

How do Kanban workers recover from failed runs like crashes or timeouts?

Kanban workers recover from failed runs by interpreting blocked, archived, timed out, crashed, or spawn-failed statuses. The worker lifecycle guidance helps them respond appropriately to these failure states and decide whether to retry or escalate.

When should a Kanban worker block a task for human review instead of completing it?

A Kanban worker should block a task for human review when dealing with code changes requiring oversight, rather than completing docs-only tasks directly. This decision relies on evaluating workspace differences and applying completion versus blocking rules.

How do I manage workspaces and tenant memory in a multi-agent Kanban system?

Managing workspaces in a multi-agent Kanban system involves understanding scratch, shared directory, and worktree behavior. Workers must apply tenant-safe memory prefixing to isolate data and ensure operational safety across different spawned runs.

Does multi-agent task management require dependency setup for worker heartbeats?

Multi-agent task management with worker heartbeats requires no external dependencies. The skill provides built-in operational requirements for summary formatting, metadata capture, and lifecycle decisions within the existing workflow.