brand-aware-design
Skill1 starsChanged 46 days agoMITView on GitHub
Design visual direction from explicit brand and product evidence, separating confirmed requirements from reversible creative decisions and unknowns. Use for brand-sensitive UI or visual work; exclude unsupported positioning and generic styling detached from evidence.
In this repository8 other files it gives its agents.
- adaptive-agent-routingSkill
- debuggingSkill
- engineering-verificationSkill
- frontend-designSkill
- human-copywritingSkill
- orchestrated-deliverySkill
- security-reviewSkill
- skill-authoringSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
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 18
Use the approved brief, brand guidelines, existing product/repository surfaces, design tokens, logo/assets, typography/color evidence, …
- Distinctive: line 30
Use this skill when a request asks to:
- Distinctive: line 37
Do not use it for:
- Distinctive: line 76
Use a compact format suited to the request, but include:
- Distinctive: line 86
Never claim “on-brand” as a fact when the brand source is incomplete.
- Distinctive: line 91
use the project's own brand-direction skill or guidelines when one exists.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Brand-aware design
Make visual decisions that are accountable to evidence. A brand-aware result is not a guessed palette, a fashionable style, or a familiar SaaS template; it is a clear record of what the source establishes, what the designer proposes, and what remains unknown.
Operational contract
Objective
Translate real brand and product evidence into a coherent visual direction without turning guesses, trends, or personal taste into brand facts. Preserve confirmed identity, make reversible creative choices explicit, and keep unknowns/conflicts visible until resolved.
Expected inputs
Use the approved brief, brand guidelines, existing product/repository surfaces, design tokens, logo/assets, typography/color evidence, imagery/iconography, product task, responsive/accessibility constraints, and any explicit user authorization to change established identity.
Required outputs
Return the evidence boundary, a compact CONFIRMED/CREATIVE DECISION/UNKNOWN/CONFLICT record, the proposed direction and rationale, do-not-change constraints, unresolved questions, responsive/accessibility implications, and implementation/review evidence actually inspected or executed.
Definition of done
The design direction is done when every material identity claim is traceable to inspected evidence or labeled as a reversible proposal, conflicts are not silently arbitrated, established brand elements are preserved unless change was authorized, and any statement such as “on-brand” is scoped to the evidence that was actually reviewed.
Trigger and exclusion boundary
Use this skill when a request asks to:
- apply or preserve a brand identity in UI, product surfaces, campaigns, or visual systems;
- review whether a design is consistent with an existing brand;
- establish a design direction from supplied brand files, repository evidence, or approved guidelines;
- adapt a known identity across responsive states, components, or channels.
Do not use it for:
- inventing brand positioning, audience, values, naming, or claims without evidence;
- generic “make it look premium/modern” work with no source or product context;
- copy-only tasks, where
human-copywritingapplies; - substantial implementation architecture or performance analysis;
- borrowing another brand's identity because the target evidence is missing.
Evidence classes
Maintain these classes in every material design decision:
| Class | Meaning | What to do |
| CONFIRMED | Directly supported by an approved brief, existing product, public file, design token, asset, or cited guideline. | Preserve it unless the user authorizes a change; cite the source. |
| CREATIVE DECISION | A reversible choice made to fill a gap or satisfy a brief. | Label it, explain the rationale, and keep an alternative available. |
| UNKNOWN | Not established by the available evidence or contradicted by sources. | Ask, use a neutral placeholder, or defer. Never silently promote it to confirmed. |
| CONFLICT | Two sources disagree or the requested change would break a confirmed constraint. | Show the conflict and ask for resolution; do not arbitrate by taste. |
A product domain, company name, existing code, or common industry pattern is not by itself a brand specification.
Operating workflow
- State the evidence boundary: files, links, approved references, and existing implementation actually inspected. Do not imply that private, remote, or secret sources were checked.
- Inventory existing identity: logo treatment, type, colors, spacing, imagery, iconography, component patterns, tone, density, motion, and responsive behavior. Record exact paths or source names.
- Build a short evidence table with CONFIRMED, CREATIVE DECISION, UNKNOWN, and any CONFLICT.
- Define the design problem and user task. A brand decision must still serve comprehension, action, accessibility, and maintainability.
- Propose the smallest coherent direction. Preserve existing identity where it is sufficient; introduce new choices only where the evidence or task requires them.
- Validate contrast, focus, keyboard behavior, content states, responsive layouts, reduced motion, and long or localized content when applicable.
- Handoff with explicit decisions, unresolved gaps, and what must not be changed without approval.
Avoid generic output
Do not use a trendy gradient, default dashboard cards, rounded containers, stock “premium” language, arbitrary type pairing, or a fashionable effect as a substitute for evidence. Do not call a direction editorial, luxury, playful, technical, organic, or minimal unless that description is confirmed or clearly labeled as a creative proposal.
Avoid flattening an existing identity into a neutral component library. If the product is operational, preserve useful density and task orientation; if it is editorial, preserve reading rhythm and hierarchy; if the source does not establish either, mark the choice UNKNOWN and propose a reversible test.
Required output
Use a compact format suited to the request, but include:
- evidence boundary and source paths;
- confirmed visual/product requirements;
- creative decisions and rationale;
- unknowns, conflicts, and questions;
- do-not-change constraints;
- accessibility and responsive implications;
- implementation or review result, with checks actually run.
Never claim “on-brand” as a fact when the brand source is incomplete. Say “consistent with the inspected evidence” and name the evidence scope.
Related skills
human-copywritingowns factual, human copy when copy is in scope.- use the project's own brand-direction skill or guidelines when one exists;
- load only the brand profile relevant to the current repository or artifact.
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.

