agentsclimarketplace

Land

Skill saadshahd/moo.md/hunch/skills/land

Use when the user wants knowledge delivered to an audience — slides, a deck, a diagram, an explainer for others, teaching or talk material — or shares existing content that isn't getting through.From its SKILL.md

Install
npx -y skills add saadshahd/moo.md --skill land

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

2.9 KB, 616 tokens by cl100k_base, as published. Nobody here has run it

Deliver information the way minds absorb it. Apply silently; explain the reasoning in plain language only when the user asks why.

Before designing

Ask only what's missing, then confirm the brief — audience and the one thing — before generating:

  1. Audience — sets language, analogies, assumed prior knowledge.
  2. The one thing — if they keep a single memory, action, or feeling, what is it? Everything serves it.

Open with a hook that creates a curiosity gap — a question, a surprising number, a misconception flipped — never a topic title.

Principles

  • One cognitive unit at a time. One idea per slide, section, or screen. Split rather than crowd.
  • Visuals carry the information. Acceptance test: delete every word — the meaning must survive. Text is a label, not the explanation. Draw the concept itself: arrow length = force, area = fraction, scene density = era.
  • Annotations live inside the graphic, attached to the structures they name — never a list beside the picture.
  • Change is shown as motion or sequence — process, relation, and comparison are animated or stepped, paced by the audience, replayable. Nothing autoplays.

Knowledge type → representation

The content is...Show it as
Concept ("what is X")An everyday analogy drawn as a graphic, correspondences mapped across (API = restaurant ordering)
Procedure ("how to")Steps revealed one at a time, prior steps stay visible; branches as a flow diagram
Relation ("X and Y")Nodes with drawn connections, Venn, or 2×2 matrix — line weight = strength
Story / historyA growing timeline, or before/after scenes where the scenery itself is the data
NumbersArea, count, or proportion anchored to a familiar referent — never a bare table
MisconceptionBefore/after flip: show the wrong belief first, then the correction

Defaults — the user's call, not yours

When the deliverable is a deck and the user hasn't specified: one self-contained HTML file; fixed canvas 1920×1080 scaled to the viewport with transform: scale() (not vw/vh); click/space/arrow advances step reveals, then slides; one coherent theme (~5 colors, 2 fonts) matched to the audience; text legible on a phone; nothing overflows the canvas.

Never

  • Bullet-point lists as the design — reach first for a layout, diagram, or step reveal, and keep the list only where a diagram would merely decorate it.
  • The narration script pasted onto the artifact — spoken words and shown visuals carry different information.
  • Dark-gradient-plus-white-text "AI deck" aesthetic.
  • Animation on a timer — a setInterval keeps firing after the slide changes, so whatever it drives outlives the slide that started it.

What ships with it: 1 file

154 B alongside SKILL.md

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.