standards-mcp
CLAUDE.md0 starsChanged 2 months agoMITView on GitHub
Every installed skill must be used whenever it is relevant — skills are not optional extras. Before producing output for any task, check whether a matching skill exists and, if so, invoke it via the Skill tool first. The full catalog of available skills (excluding DevExpress) is in SKILLS.md. - Check first, then act. For any task, match it to a skill in SKILLS.md and load that skill before doing the work from memory. - Design/UI/frontend → the design skills…
In this repository81 other files it gives its agents.
- agenticSkill
- antSkill
- artisticSkill
- basicSkill
- bentoSkill
- boldSkill
- brandkitSkill
- brutalismSkill
- cafeSkill
- claudeSkill
- claymorphismSkill
- cleanSkill
- codexSkill
- colorfulSkill
- contemporarySkill
- corporateSkill
- cosmicSkill
- creativeSkill
- design-taste-frontendSkill
- design-taste-frontend-v1Skill
- ditheredSkill
- doodleSkill
- dramaticSkill
- editorialSkill
- enterpriseSkill
- expressiveSkill
- fantasySkill
- fictionSkill
- flatSkill
- friendlySkill
- full-output-enforcementSkill
- futuristicSkill
- geometricSkill
- glassmorphismSkill
- gpt-tasteSkill
- gradientSkill
- high-end-visual-designSkill
- imagegen-frontend-mobileSkill
- imagegen-frontend-webSkill
- image-to-codeSkill
- immersiveSkill
- impeccableSkill
- industrial-brutalist-uiSkill
- levelsSkill
- lingoSkill
- materialSkill
- matrixSkill
- minimalist-uiSkill
- minimalSkill
- modernSkill
- monoSkill
- neobrutalismSkill
- neonSkill
- neumorphismSkill
- pacmanSkill
- paperSkill
- perspectiveSkill
- powerSkill
- premiumSkill
- professionalSkill
- The first 60 shown.
The file
We split this CLAUDE.md into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
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 5
Every installed skill must be used whenever it is relevant — skills are not optional extras.
- Distinctive: line 45
Before any UI/frontend work is considered done, it must be reviewed with the web-design-guidelines skill (Vercel Web Interface Guidelines).
- Distinctive: line 47
Use it to review UI code, check accessibility, audit design/UX, and verify compliance with web best practices.
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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
Project rules
Skills usage (MANDATORY — applies to ALL tasks)
Every installed skill must be used whenever it is relevant — skills are not optional extras. Before producing output for any task, check whether a matching skill exists and, if so, invoke it via the Skill tool first. The full catalog of available skills (excluding DevExpress) is in SKILLS.md.
Rules:
- Check first, then act. For any task, match it to a skill in
SKILLS.mdand load that skill before doing the work from memory. - Design/UI/frontend → the design skills are mandatory (see the Design section below).
- Documents →
docx/pdf/pptx/xlsxfor those file types;slidesfor presentations. - Charts / dashboards / any data visualization →
datavizbefore writing chart code. - Animation / motion →
animate(web) oranimate-expo(React Native/Expo). - Code review / cleanup / security →
code-review,simplify,security-review. - Artifacts (published web pages) →
artifact-design, andartifact-capabilities/artifact-diagrammingas needed. - Anything else with a matching skill → use it. If unsure whether a skill fits, prefer using it over guessing.
- Skills complement, they do not replace, the engineering standards (standards 01–14) or the mandatory design review gate.
Design (MANDATORY)
Any design, UI, frontend, or visual work in this project must apply the taste-skills installed under .agents/skills/. This is not optional — invoke the relevant skill via the Skill tool before producing design output.
Available design skills and when to use each:
- design-taste-frontend — default for landing pages, portfolios, redesigns, and any non-templated UI. Use this first for most frontend/design tasks.
- design-taste-frontend-v1 — only when exact backward-compatible v1 behavior is explicitly required.
- high-end-visual-design — premium/high-end visual design work.
- minimalist-ui — minimalist interface styling.
- industrial-brutalist-ui — brutalist/industrial aesthetic.
- stitch-design-taste / gpt-taste — design taste references for taste calibration.
- redesign-existing-projects — when redesigning or restyling existing code (audit-first).
- brandkit — brand guidelines, logo systems, identity decks, visual-world presentations.
- image-to-code — converting a design image/mockup into code.
- imagegen-frontend-web / imagegen-frontend-mobile — AI image generation for web/mobile frontends.
- full-output-enforcement — enforce complete, non-truncated output on design deliverables.
Rule of thumb: pick the skill(s) that match the task, read them, and follow their guidance. Never ship design/UI output in this project without going through the appropriate taste-skill.
Web Interface Guidelines review (MANDATORY)
Before any UI/frontend work is considered done, it must be reviewed with the web-design-guidelines skill (Vercel Web Interface Guidelines). This is a required gate, not optional:
- Use it to review UI code, check accessibility, audit design/UX, and verify compliance with web best practices.
- The skill fetches the latest guidelines from
https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.mdand checks the target files against every rule, reporting findings infile:lineformat. - Apply it both when building new UI (review the result) and when redesigning existing UI.
Combined flow for design/UI tasks in this project: build with the matching taste-skill → review with web-design-guidelines → fix findings before shipping.
Design-system style skills (use as needed)
A library of ~67 design-system / visual-style skills (from bergside/awesome-design-skills) is installed under .agents/skills/. Use them whenever a task calls for a specific visual direction or aesthetic — pick the one(s) that match the requested look and apply their tokens, components, and guidance.
Available styles:
- Minimal / clean: minimal, clean, basic, flat, mono, refined, sleek, spacious, square, impeccable
- Modern / professional: modern, contemporary, professional, corporate, enterprise, premium, editorial, shadcn, material
- Bold / expressive: bold, brutalism, neobrutalism, expressive, dramatic, power, vibrant, colorful, gradient, neon, cosmic, futuristic, immersive
- Textured / depth: glassmorphism, claymorphism, neumorphism, skeumorphism, perspective, levels, pulse
- Artistic / playful: artistic, creative, doodle, sketch, dithered, riso, paper, friendly, storytelling, geometric, bento
- Retro / themed: retro, vintage, cafe, terracotta, fantasy, fiction, sega, roku, matrix, pacman, tetris
- Agent/tool-native: agentic, claude, codex, stitch, lingo, ant
Rule of thumb: if the user names or implies a style (e.g. "make it glassmorphic", "brutalist landing page", "clean SaaS dashboard"), load the matching style skill first. These complement — not replace — the mandatory taste-skill build step and the web-design-guidelines review gate above.
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.

