kanban-worker

Standardizes Hermes kanban worker task handoffs and resolves common workflow errors.

3|Updated Apr 21, 2026
One-click install
npx skills add https://github.com/DarkArty07/Aether-Agents --skill kanban-worker-darkarty07
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kanban-worker
Source: https://github.com/DarkArty07/Aether-Agents/tree/main/home/skills/devops/kanban-worker
Command: npx skills add https://github.com/DarkArty07/Aether-Agents --skill kanban-worker-darkarty07

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Hermes kanban workers frequently encounter avoidable errors like stale workspace conflicts, phantom task claims, context leaks across tenants, and failed cross-agent handoffs that waste compute and break multi-agent task pipelines.

Core Features & Use Cases

  • Validated workspace handling: Clear rules for working in scratch, shared persistent directory, and git worktree environments to prevent state conflicts between runs.
  • Standardized task handoffs: Pre-built kanban_complete metadata patterns for coding, research, and review tasks that let downstream workers parse results without re-reading full prose.
  • Edge case guardrails: Guidance for retry scenarios, block reason formatting, heartbeat best practices, and tenant isolation to prevent context leaks and failed runs. Use case: A security worker reviewing a pull request can use the structured handoff format to pass critical findings to remediation workers, and avoid claiming task IDs they did not create to prevent workflow errors.

Quick Start

Invoke the kanban-worker skill when dispatched as a Hermes kanban worker to follow validated best practices for task completion, handoffs, and error handling.

Frequently Asked Questions about kanban-worker

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

FAQPage Schema
How do I prevent context leaks and phantom task claims in multi-agent kanban worker workflows?

To prevent context leaks and phantom task claims in multi-agent kanban worker workflows, apply tenant isolation guardrails and standardized task handoff metadata to ensure workers only claim tasks they created and maintain strict execution boundaries.

What is the best way to format cross-agent task handoffs for delegated coding and research tasks?

The best way to format cross-agent task handoffs is using pre-built kanban_complete metadata patterns for coding, research, and review tasks, allowing downstream workers to parse results without re-reading full prose.

How do I handle workspace state conflicts between runs in scratch, shared directories, or git worktrees?

Handle workspace state conflicts by following validated workspace handling rules for scratch, shared persistent directories, and git worktrees, which prevent stale workspace conflicts and state corruption between multi-agent runs.

Why do cross-agent handoffs fail in Hermes kanban worker pipelines and how can I fix them?

Cross-agent handoffs fail due to unstructured output and missing metadata, but you can fix them by applying standardized handoff formats and retry diagnostics to ensure downstream agents successfully parse task results.

Does this approach support retry diagnostics and block reason formatting for failed delegated tasks?

Yes, this approach supports retry scenarios and block reason formatting by providing edge case guardrails for failed runs, ensuring clear diagnostic output and heartbeat best practices for multi-agent task pipelines.

When do I need tenant isolation for multi-agent task handoffs in shared workspaces?

You need tenant isolation for multi-agent task handoffs when multiple workers operate in shared workspaces, preventing context leaks across tenants and ensuring secure, separated execution environments for delegated tasks.