genre-picker

Classify knowledge artifacts into skill, pattern, runbook, template, or spec meta-types before creation.

Updated Nov 1, 2024
One-click install
npx skills add https://github.com/mlorentedev/dotfiles --skill genre-picker-mlorentedev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: genre-picker
Source: https://github.com/mlorentedev/dotfiles/tree/main/harness/skills/genre-picker
Command: npx skills add https://github.com/mlorentedev/dotfiles --skill genre-picker-mlorentedev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Knowledge artifacts in a 00_meta knowledge store often get filed in the wrong folder, causing drift that later requires cleanup sweeps. This Skill applies boundary tests at write-time so every new artifact lands in the correct meta-type folder from the start. ## Core Features & Use Cases - Ordered boundary tests: Applies five sequential tests (trigger-based, ordered steps, principle, skeleton, single-change scope) to determine whether knowledge is a skill, runbook, pattern, template, or spec. - Store/repo/forge discriminator: First decides whether knowledge belongs in the cross-project store, a single repo's docs, or GitHub issues, preventing misplaced artifacts entirely. - Pattern vs runbook disambiguation: Resolves the blurry line between principles applied with judgment and step-by-step procedures, with a reclassify-never-duplicate rule. - Use Case: Before writing a new note about onboarding any repo to a placement model, run the classification to confirm it is a cross-project runbook and place it in 00_meta/runbooks/ rather than patterns. ## Quick Start Ask the assistant to classify whether a piece of knowledge is a pattern or a runbook before creating the file in 00_meta.

Frequently Asked Questions about genre-picker

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

FAQPage Schema
How do I decide if knowledge is a pattern or a runbook?

A pattern states a principle like prefer X over Y because Z and is applied with judgment, while a runbook is an ordered sequence of steps executed start to finish. If the text is mostly a checklist run top to bottom, it is a runbook; if it mostly explains why, it is a pattern.

How do I classify a new knowledge artifact before creating it?

Apply the boundary tests in order: trigger-based invocation means skill, ordered cross-project steps mean runbook, a principle means pattern, a blank skeleton means template, and single-change scope with acceptance criteria means spec. The first matching test wins.

When should knowledge go to a repo instead of the 00_meta store?

Project-bound knowledge such as an ADR, lesson, or deployment procedure for one repository belongs in that repo's docs folder, not the cross-project store. Only knowledge spanning multiple repos qualifies for 00_meta.

What should I do with an artifact that is filed in the wrong folder?

Re-classify it using the boundary tests and move it to the correct folder. Never duplicate the artifact in two locations; the rule is reclassify, never duplicate.

What are the limitations of this classification approach?

It only covers cross-project knowledge destined for the 00_meta store and defers the authoritative taxonomy to the pattern-knowledge-placement reference. Tactical tasks and discussions belong in GitHub issues and are out of scope.