agent-instructions
Use when writing project instructions for a coding agent (CLAUDE.md, AGENTS.md, or equivalent). Covers what belongs in them, what does not, structure, and…
Use when applying or documenting a brand identity. Covers extracting a system from existing assets, documenting colour, type, and voice, and applying a brand consistently across new work.
$ npx -y skills add nimadorostkar/Claude-Skills-collection --skill brand-guidelines --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/brand-guidelinesContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when applying or documenting a brand identity. Covers extracting a system from existing assets, documenting colour, type, and voice, and applying a brand consistently across new work.
name: brand-guidelines description: Use when applying or documenting a brand identity. Covers extracting a system from existing assets, documenting colour, type, and voice, and applying a brand consistently across new work. metadata: category: design version: 1.0.0 tags: [brand, identity, guidelines, consistency, style]
Apply a brand consistently, and document it well enough that someone else can too. A brand that exists only in one designer's judgment is not a brand; it is a taste.
1. **Extract before inventing** — If a brand exists in assets but not in a document, derive it: sample the colours, identify the typefaces, measure the spacing, read the copy for voice. What is consistent across the assets is the brand; what varies is drift. 2. **Document the decisions, not just the values** — "Primary blue #2563EB, used for actions and links only. Never for large fills." The constraint is the useful part. 3. **Cover the cases people get wrong** — Minimum logo size, clear space, what to do on a dark background, what happens when the brand colour fails contrast. 4. **Document the voice** — It is half the brand and it is nearly always omitted. Give examples of on-brand and off-brand copy. 5. **Apply, then audit** — Check the output against the system rather than against instinct.
**A brand system documented so that it can be applied:**
## Colour | Token | Hex | Use | Never | |----------------------|----------|------------------------------------------|------------------------------| | brand/primary | #2563EB | Primary actions, links, active states | Large fills; body text | | brand/primary-dark | #1D4ED8 | Hover and pressed states only | Anything else | | brand/ink | #111827 | Body text, headings | — | | brand/paper | #FFFFFF | Backgrounds | — | | brand/accent | #F59E0B | Warnings, highlights. Sparingly. | More than one per screen | **Contrast:** `brand/primary` on white is 4.6:1 — it passes for body text but only just. For text below 16px, use `brand/primary-dark` (7.1:1). Never place `brand/accent` text on white: it is 2.1:1 and unreadable. ## Type Display : Söhne, 600 weight. Headings only. Tracking -0.02em above 32px. Text : Söhne, 400/500. Body, UI, everything else. Never : a third typeface. Hierarchy comes from size and weight. ## Voice We are direct and specific. We do not hedge, and we do not perform enthusiasm. | Off-brand | On-brand | |----------------------------------------------------|---------------------------------------------| | "Oops! Something went wrong." | "We couldn't save your changes. Try again." | | "We're thrilled to announce an exciting new..." | "Exports now include line items." | | "Please kindly note that you may wish to consider" | "You can also..." | ## Logo - Minimum width: 24px (digital), 20mm (print). - Clear space: the height of the mark, on all four sides. - On photography: use the white version, on an area of the image with low detail. Never on a busy background. - Never: stretch, rotate, recolour, add effects, or place inside a shape that is not part of the mark.
A curated library of 137 production-grade skills for Claude and other AI coding agents. Every skill follows one structure, speaks with one voice, and earns its place by changing what the agent does.
Repo: nimadorostkar/Claude-Skills-collection
Use when writing project instructions for a coding agent (CLAUDE.md, AGENTS.md, or equivalent). Covers what belongs in them, what does not, structure, and…
Use when an agent needs state that survives a session or a context compaction. Covers what to persist, file-based memory, structuring notes for retrieval, and…
Use when automating agent behavior with lifecycle hooks. Covers hook events, deterministic enforcement of rules the model should not be trusted to remember,…
Use when packaging skills, commands, hooks, and MCP servers into a distributable plugin. Covers manifest structure, bundling, versioning, testing, and…
Use when writing a new skill for an AI agent. Covers scoping, description writing for reliable triggering, progressive disclosure, and the difference between a…
Use when reviewing or improving an existing agent skill. Covers triggering accuracy, content quality, redundancy with the base model, and measuring whether the…