kanban-worker

Guide Hermes Kanban worker runs on workspace handling, tenant isolation, and card claiming.

Updated May 26, 2026
One-click install
npx skills add https://github.com/ruiyangruiyi/hermes-agent --skill kanban-worker-ruiyangruiyi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kanban-worker
Source: https://github.com/ruiyangruiyi/hermes-agent/tree/main/skills/devops/kanban-worker
Command: npx skills add https://github.com/ruiyangruiyi/hermes-agent --skill kanban-worker-ruiyangruiyi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents common failures in Hermes Kanban worker runs by documenting correct workspace handling, safe cross-tenant memory usage, and reliable handoff/claiming practices.

Core Features & Use Cases

  • Workspace handling guidance: Explains how to behave for scratch, dir:<path>, and worktree workspaces to avoid stale or invalid file assumptions.
  • Tenant isolation rules: Ensures persistent memory writes are safely namespaced under $HERMES_TENANT to prevent context leakage.
  • Correct tool handoffs: Shows how to structure kanban_complete summaries and metadata for downstream parsers, including retry diagnostics and heartbeats.
  • Safe card claiming: Enforces using only kanban_create return values when reporting created_cards, preventing phantom or cross-worker claims.
  • Fast triage for blocks and retries: Recommends specific kanban_block reasons and how to interpret prior run outcomes when retrying.

Quick Start

Load the kanban-worker Skill and start by calling kanban_show to confirm the task is not blocked or archived before you proceed with your work.

Frequently Asked Questions about kanban-worker

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

FAQPage Schema
How do I prevent worker errors in distributed kanban task execution?

To prevent worker errors in distributed kanban task execution, follow workspace handling guidance for scratch, dir:<path>, and worktree environments to avoid stale file assumptions and ensure reliable task handoffs.

What is the correct way to handle tenant isolation in multi-agent kanban workflows?

Tenant isolation in multi-agent kanban workflows requires namespacing persistent memory writes safely under $HERMES_TENANT to prevent context leakage across different workers and tenants during execution.

How do I safely claim kanban cards created by other workers?

To safely claim kanban cards, use only the return values captured from kanban_create when reporting created_cards, which prevents phantom or cross-worker claims that disrupt multi-agent orchestration.

How do I structure kanban_complete metadata for downstream retry diagnostics?

Structure kanban_complete metadata for retry diagnostics by producing parseable completion summaries and sending timely heartbeats, enabling accurate interpretation of prior run outcomes during subsequent execution attempts.

When should I use kanban_block in distributed task handling?

Use kanban_block in distributed task handling when encountering specific execution obstacles, providing recommended block reasons to enable fast triage and accurate retry diagnostics for subsequent workflow attempts.

Does kanban-worker require specific workspace handling for different workspace types?

Yes, kanban-worker requires specific workspace handling behaviors for scratch, dir:<path>, and worktree workspaces to avoid stale or invalid file assumptions during distributed task execution.