design-ontology
AGENTS.md0 starsChanged 2 days agoMITView on GitHub
You are in design-ontology: design styles as data, compiled into agent skills.
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository19 other files it gives its agents.
- bauhausSkill
- bento-grid-uiSkill
- brutalist-architectureSkill
- corporate-memphisSkill
- flat-designSkill
- frutiger-aeroSkill
- glassmorphismSkill
- grunge-editorialSkill
- international-style-architectureSkill
- material-designSkill
- memphis-designSkill
- mid-century-modernSkill
- neo-brutalism-uiSkill
- neumorphismSkill
- psychedelicSkill
- skeuomorphismSkill
- spatial-visionosSkill
- swiss-internationalSkill
- y2k-aestheticSkill
The file
We split this AGENTS.md into the rules it gives an agent
4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 14
Do not edit skills/, graph/ or the lineage block in README.md by hand.
- Very specific: line 15
After each change to the data, run npm run build.
- Very specific: line 16
Before you commit, run npm run typecheck if you changed a script.
- Very specific: line 17
Commit the data and the generated files together.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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
Instructions for agents
You are in design-ontology: design styles as data, compiled into agent skills.
To use a style
- Read
skills/SKILL_INDEX.md. - Pick one row. Load
skills/<id>/SKILL.md. Load one skill at a time. - Obey its "System directives" and "Guardrails". If the user asks for a variation, follow its "Lineage fallbacks".
To change the repository
- The source of truth is
data/<domain>/styles/*.json,data/<domain>/eras/*.jsonanddomains/*.json. - Do not edit
skills/,graph/or the lineage block inREADME.mdby hand. They are generated. - After each change to the data, run
npm run build. It must report 0 errors. - Before you commit, run
npm run typecheckif you changed a script. - Commit the data and the generated files together. The pre-commit hook does this for you and moves the package version one patch step. Do not edit the patch number by hand; do not skip it with
--no-verify. - In a page, set text with the derived tokens:
--color-on-bgon the background,--color-on-surfaceon a surface,--color-on-primaryon a primary button. Do not put--color-texton a surface. - Each skill has a visual pack in
skills/<id>/pack/:hero.svg,pattern.svg,icons.svg. Use them before you draw your own. The sprite is an open-source stand-in: for a product, use the icon library of the brand and choose the family by the rules in the skill. - The content rules are in
CONTRIBUTING.md. Read them before you write a style.
Commands
| Command | Result |
npm run build | validate, build the graph, compile the skills |
npm run validate -- --schema-only | schema check only; use it while a batch of styles refers to files that are not yet present |
npm run check | build, then fail if a generated file differs from the commit (CI runs this) |
npm run query -- list | list all styles with their vectors |
npm run query -- near <id> | nearest styles by vector distance |
npm run query -- path <a> <b> | shortest lineage path |
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.