yak-shaving-handbook

Guide shavers through the yakthang task lifecycle and messaging workflow.

3|3|Updated Feb 13, 2026
One-click install
npx skills add https://github.com/mrdavidlaing/yakthang --skill yak-shaving-handbook
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: yak-shaving-handbook
Source: https://github.com/mrdavidlaing/yakthang/tree/main/skills/yak-shaving-handbook
Command: npx skills add https://github.com/mrdavidlaing/yakthang --skill yak-shaving-handbook

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Yak Shaving Handbook provides a comprehensive operating guide for shavers in the yakthang environment, clarifying how tasks are created, progressed, blocked, and completed using yx, context.md, comments.md, and shaver-message.

Core Features & Use Cases

  • Task lifecycle and deepest-first execution model, showing how parents are blocked by children.
  • Guidance on message channels and fields (context.md, comments.md, shaver-message, yakob-message) to communicate progress and decisions.
  • Heartbeat and session discipline overview using /loop to schedule status checks and ensure bounded work sessions.

Quick Start

Read this handbook to learn the Yakob–Yak task lifecycle, messaging channels, and heartbeat workflow so you can start shaving tasks correctly.

Frequently Asked Questions about yak-shaving-handbook

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

FAQPage Schema
How do I coordinate task lifecycles across multiple agents?

Task lifecycle coordination across multi-agent tasks requires formalizing frontmatter metadata and applying a deepest-first execution model where parent tasks are blocked by child tasks until completion.

How does heartbeat coordination work for live progress signals?

Heartbeat coordination uses scheduled /loop status checks to provide live progress signals and enforce bounded work sessions, ensuring continuous task tracking across the multi-agent environment.

What is the deepest-first execution model for task workflows?

The deepest-first execution model processes nested tasks by blocking parent tasks until all child tasks are completed, ensuring dependencies are resolved before parent workflows advance.

Which message channels should I use to communicate task progress and decisions?

Task progress and decisions are communicated through context.md, comments.md, shaver-message, and yakob-message channels, each handling specific messaging interactions within the lifecycle.

Can I use this workflow to manage blocked and completed task states?

Yes, the workflow formally defines how tasks are created, progressed, blocked, and completed using yx and documented conventions, providing structured state management across the environment.

Do I need specific frontmatter metadata to start shaving tasks correctly?

Yes, formalized frontmatter metadata is required to adhere to lifecycle conventions and messaging interactions, enabling proper task creation and progression within the environment.