god-mode
Skill1 starsChanged 20 days agoMITView on GitHub
Use when someone wants a website, web app, landing page, CRM, dashboard, SaaS product, portfolio or browser game to look significantly better, more premium or less "AI-generated". Runs the full pipeline (read the project, score it, pick a direction, upgrade, polish, verify) and routes to the focused design-god-mode skills. Not for backend-only work.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- GeekFamilyCorp/TricorderKittheir own version, first seen Oct 13, 2026
- contactmridulhere-commits/claude-skillstheir own version, first seen Sep 27, 2026
- AbuBakarPial/universal-god-modetheir own version, first seen Jun 1, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- GeekFamilyCorp/TricorderKit★ 32same name
- AbuBakarPial/universal-god-mode★ 4same name
- contactmridulhere-commits/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 4 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
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
- Distinctive: line 12
Read only what you need: the layout/root component, global CSS or Tailwind config, the theme file, and the 2-3 pages the user cares about.
- Very specific: line 13
Run node <skill-base-dir>/../audit/scripts/slop-scan.mjs <project-dir> before reading styles.
- Distinctive: line 16
Look at the result at most twice per page (desktop 1440px, phone 390px).
- Very specific: line 38
Use the audit skill. Get a score out of 100 and the ranked fix list.
- Very specific: line 41
Use the direction skill. It produces a short token plan (colours, type, spacing, radius, shadow, motion) and checks it against generic …
- Distinctive: line 43
If the brief is ambiguous on one thing that changes everything (for example "keep the brand colours or not?"), ask that one question.
- Very specific: line 51
Use the polish skill. Then look at the rendered pages.
- Distinctive: line 60
Add a heavy dependency for looks alone.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Claude Design God Mode
You are the design director and the design engineer in one. The job is a visibly better interface that still works exactly as before. Ship finished work, not suggestions.
Budget rules (read first, they keep this cheap)
- Read only what you need: the layout/root component, global CSS or Tailwind config, the theme file, and the 2-3 pages the user cares about. Do not read every component.
- Run
node <skill-base-dir>/../audit/scripts/slop-scan.mjs <project-dir>before reading styles. Its summary replaces reading most CSS by hand. - Change the system before the screens: tokens, fonts and global styles first. One token edit fixes fifty components.
- Edit in place with small diffs. Never rewrite a whole file to change its look.
- Look at the result at most twice per page (desktop 1440px, phone 390px). Batch every fix from one look before looking again.
- Load a reference file only when its step needs it.
Pipeline
1. Read the room (2-5 tool calls)
Find: the stack (framework, CSS approach, component library), the product, the audience, the main pages, and any existing brand (logo colours, fonts, DESIGN.md). Write a one-line Design read:
<product> for <audience>. Mode: <mode>. Keep: <what must survive>. Stack: <stack>.
2. Pick the mode
The mode is set by the page, not the company.
| Mode | The visitor wants to | Examples | Ambition shows as |
| Persuade | decide and act | landing, pricing, marketing, local business site | one bold idea, real imagery, sharp copy |
| Operate | finish a task fast | CRM, dashboard, admin, settings, forms | density, clarity, calm colour, precise detail |
| Play | feel something | game menus, HUD, interactive toys | juice, feedback, legibility in motion |
| Read | understand | docs, blog, help centre | type, rhythm, navigation |
A CRM's marketing site is Persuade. Its customer list is Operate.
3. Score it
Use the audit skill. Get a score out of 100 and the ranked fix list. Skip this only when the user asked for one narrow change.
4. Choose the direction
Use the direction skill. It produces a short token plan (colours, type, spacing, radius, shadow, motion) and checks it against generic defaults. If the project already has a strong brand, refine it; do not replace it.
If the brief is ambiguous on one thing that changes everything (for example "keep the brand colours or not?"), ask that one question. Otherwise decide and say what you decided.
5. Upgrade
- Existing site: use the
redesignskill (safe order, nothing breaks). - Then the mode skill:
landing(Persuade),app-ui(Operate),game-ui(Play). Read mode usesdirectiontype rules pluspolish. - Motion only where it helps:
motion.
6. Polish and verify
Use the polish skill. Then look at the rendered pages. If you cannot run the app, say so plainly and list what you could not check.
7. Report
Finish with: before score, after score, the 3-6 biggest changes in plain words, and anything the user must supply (real photos, logo files, copy facts). Do not claim a score you did not re-measure.
Never
- Break behaviour: routes, forms, handlers, data, tests, SEO tags and analytics stay intact.
- Invent facts: no fake testimonials, customer logos, stats or awards. Use clearly marked placeholders.
- Add a heavy dependency for looks alone. Prefer CSS. Ask before adding a UI library.
- Override a pinned brief. If the user names a colour, font or era, follow it.
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.