enlighten
Skill7 starsChanged 3 months agoMITView on GitHub
Explain a topic to someone with no background, as a self-contained HTML picture explainer. Use when the user types /enlighten <topic> or asks for a dead-simple picture explanation of how something works.
In this repository25 other files it gives its agents.
- .github/copilot-instructions.mdCopilot instructions
- brainstormingSkill
- ce-brainstormSkill
- ce-compound-refreshSkill
- ce-compoundSkill
- ce-ideateSkill
- ce-planSkill
- ce-reviewSkill
- ce-workSkill
- claude-permissions-optimizerSkill
- deepen-planSkill
- document-reviewSkill
- evolveSkill
- feature-videoSkill
- instinctsSkill
- karpathy-guidelinesSkill
- learnSkill
- lfgSkill
- observeSkill
- ralph-loopSkill
- resolve_todo_parallelSkill
- setupSkill
- slfgSkill
- test-browserSkill
- unslopSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
If the topic is a repository or a path, read the source first and explain what is actually there.
- Distinctive: line 25
Keep the interesting parts.
- Distinctive: line 28
If the source states limits, bugs, or gaps, say so plainly at the end.
- Distinctive: line 53
Do not trust a screenshot for layout.
- Distinctive: line 59
Check at a narrow width (375px) and a wide one.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
enlighten
Explain like I'm someone who knows nothing about this topic, using a self-contained HTML page with big pictures and few words per idea.
Topic: $ARGUMENTS
This is the one verbose mode in this repo. Everything else here cuts words. Enlighten spends them, because a reader who does not know the topic cannot expand a fragment.
If the topic is a repository or a path, read the source first and explain what is actually there. Never explain from the name alone.
Write it for someone with no background
- Short sentences. Everyday words. If a term needs defining, use a different term.
- One idea per section, with a picture for it.
- Concrete analogies over abstractions: a bus, a thermostat, a locked door.
- Keep the interesting parts. Simple does not mean shallow. The clever detail is usually the reason the thing exists, so find a plain-language version of it rather than dropping it.
- If the source states limits, bugs, or gaps, say so plainly at the end. An explainer that only flatters is not an explanation.
- Still no filler: no preamble, no closing recap.
Layout rules, so the artifact does not break
Text bleeding out of its box is the failure mode of this skill. It happens when labels are positioned by hand. Avoid it structurally:
- Lay out with HTML and flexbox, never with hand-placed coordinates. Use
display:flex; flex-wrap:wraprows of boxes. Text then wraps and boxes size to their contents on any screen. - Use SVG only for pure geometry with no text in it: circles, arrows, bars. The moment a diagram needs a label, put the label in HTML next to or beneath the SVG.
- Never put
<text>inside SVG. SVG text does not wrap, and its width depends on whichever font the viewer has installed, so every hand-placed label is a guess that breaks on someone else's machine. - Emoji and unicode arrows make good picture elements and cost nothing.
- Cap prose at about
max-width: 46chso lines stay readable. - Give the page a mobile breakpoint: shrink the glyphs, and rotate horizontal arrows 90 degrees when the row stacks into a column.
Verify before saying it is done
Do not trust a screenshot for layout. Serve the file, then measure:
- Compare
document.documentElement.scrollWidthagainstclientWidth. They must match, or something is overflowing horizontally. - For each element, check
getBoundingClientRect().rightagainst its parent's right edge. Nothing should exceed it. - Check at a narrow width (375px) and a wide one.
Report the measurements. A screenshot can be cropped by the capture tool and show clipping that is not really there, which reads as a bug you then chase for nothing.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

