onboard

onboard team members to Trellis workflows with concepts, walkthroughs, and spec customization.

92|8|Updated May 15, 2026
One-click install
npx skills add https://github.com/LonelyHerbivore/Trellis-Herbivore --skill onboard-lonelyherbivore
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: onboard
Source: https://github.com/LonelyHerbivore/Trellis-Herbivore/tree/main/packages/cli/src/templates/codex/skills/onboard
Command: npx skills add https://github.com/LonelyHerbivore/Trellis-Herbivore --skill onboard-lonelyherbivore

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the onboarding challenge for new contributors who need to quickly understand how Trellis manages context, project-specific knowledge, and quality gates so they can collaborate effectively with an AI coding workflow.

Core Features & Use Cases

  • Core Concepts for AI-assisted development: Explains why Trellis uses workspace memory, spec injection, and check/finish gates to address context loss, generic model behavior, and context drift.
  • Real-world workflow walkthroughs: Teaches how to apply the skills in five practical scenarios (bug fix, planning-only, code review fixes, large refactoring, and debugging) with explicit “principle / what happens / if skipped” reasoning.
  • Guideline customization workflow: Helps confirm whether .trellis/spec/ contains empty templates and guides the developer to fill project conventions so $before-*-dev becomes genuinely useful.

Quick Start

Use the onboard skill when a new developer joins or when Trellis guidelines in .trellis/spec/ need initial setup, then follow the three parts to learn the workflow and customize the spec templates.

Frequently Asked Questions about onboard

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

FAQPage Schema
How do I set up the Trellis spec guidelines for a new AI-assisted coding project?

To set up Trellis spec guidelines, you must verify the `.trellis/spec/` directory contains empty templates and customize them with project conventions. This ensures `$before-*-dev` injections provide genuinely useful context for AI workflows.

How does workspace memory prevent context drift in AI development workflows?

Workspace memory prevents context drift by maintaining persistent project-specific knowledge across AI coding sessions. Trellis uses this alongside spec injection and quality gates to address context loss and generic model behavior during development.

What is the best way to onboard new developers to an AI-assisted workflow with code review gates?

The best way to onboard developers is using a three-part structure: explaining core workspace memory concepts, demonstrating real-world workflow walkthroughs for bug fixes and refactoring, and customizing development guideline templates.

Which real-world scenarios should I walkthrough to understand AI-assisted refactoring and debugging?

You should walkthrough five practical scenarios: bug fixes, planning-only tasks, code review fixes, large refactoring, and debugging. Each scenario provides explicit principle, execution, and consequence reasoning for the AI workflow.

Does Trellis workflow support guideline customization for existing project conventions?

Yes, Trellis workflow supports guideline customization by checking if `.trellis/spec/` contains empty templates and guiding developers to fill them with project conventions. This prepares reliable future runs with properly injected context.