kanban-worker

Validate Hermes Kanban worker runs and structure kanban_complete handoffs.

19|4|Updated Apr 22, 2026
One-click install
npx skills add https://github.com/carterwayneskhizeine/hermes-agent-windows-R --skill kanban-worker-carterwayneskhizeine
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kanban-worker
Source: https://github.com/carterwayneskhizeine/hermes-agent-windows-R/tree/main/skills/devops/kanban-worker
Command: npx skills add https://github.com/carterwayneskhizeine/hermes-agent-windows-R --skill kanban-worker-carterwayneskhizeine

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents failed or inconsistent Kanban worker runs by clarifying workspace behavior, tenant isolation, completion handoff shapes, and retry/block edge cases.

Core Features & Use Cases

  • Workspace handling & isolation: Guides how to operate in scratch, persistent dir, and git worktree environments without corrupting shared state.
  • Completion handoff correctness: Defines how to structure kanban_complete summary and metadata for downstream parsing, plus how to correctly claim created_cards using real kanban_create return values.
  • Operational safety for retries and blocks: Instructs workers to validate current task state with kanban_show, interpret run outcomes, and use kanban_block with actionable reasons instead of blind retries.

Quick Start

Ask an AI agent to complete its assigned Hermes Kanban card using kanban_show first, follow KANBAN_GUIDANCE for the lifecycle, and report a precise kanban_complete(summary, metadata, created_cards) handoff using the correct workspace and retry semantics.

Frequently Asked Questions about kanban-worker

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

FAQPage Schema
How do I ensure correct Kanban worker handoffs when running tasks across different workspaces?

Ensure correct Kanban worker handoffs by structuring kanban_complete outputs with summary, metadata, and created_cards. Validate workspace behavior across scratch, persistent directories, or git worktrees to prevent shared state corruption during multi-agent operations.

Why do my Kanban retry attempts fail or create phantom card claims?

Kanban retry attempts fail when workers blindly retry without validating current task state. Use kanban_show to interpret run outcomes and kanban_block with actionable reasons to prevent phantom or unauthorized card claims during edge case handling.

What is the correct way to claim created cards in a multi-tenant Kanban environment?

Claim created cards in a multi-tenant Kanban environment by using real kanban_create return values. This ensures tenant isolation and prevents unauthorized claims, maintaining memory integrity across dispatched worker tasks.

How do I format kanban_complete metadata for downstream parsing?

Format kanban_complete metadata for downstream parsing by providing a structured summary, precise metadata, and created_cards. This structured output ensures downstream processes can correctly interpret and route the completed Kanban task results.

When should I use kanban_block instead of retrying a failed Kanban task?

Use kanban_block instead of retrying when a Kanban task encounters an actionable error. Validate the run outcome with kanban_show first, then provide a specific reason in kanban_block to halt execution safely rather than forcing blind retries.

Does this Kanban lifecycle guidance work for git worktree environments?

Yes, this Kanban lifecycle guidance works for git worktree environments. It instructs workers on safe operational behavior across scratch, persistent shared directories, and worktrees to avoid corrupting shared state during multi-agent task execution.