visual-assets-design
Skill0 starsChanged 25 days agoMITView on GitHub
Specify, produce where tools permit, and quality-check visual assets for apps, social and print with geometry, accessibility, typography, licensing and honest source-file delivery.
In this repository28 other files it gives its agents.
- agent-skill-security-auditorSkill
- backend-api-engineerSkill
- backend-debuggingSkill
- backup-disaster-recoverySkill
- code-review-prSkill
- database-transaction-integritySkill
- dependency-supply-chain-auditorSkill
- devops-ci-cdSkill
- documentation-handoverSkill
- frontend-design-directionSkill
- llm-api-integrationSkill
- load-resilience-testingSkill
- mcp-integration-builderSkill
- observability-monitoringSkill
- performance-accessibilitySkill
- playwright-webapp-testingSkill
- privacy-data-governanceSkill
- product-business-analystSkill
- project-bootstrap-environmentsSkill
- qa-release-gatekeeperSkill
- qa-test-strategySkill
- react-artifacts-builderSkill
- security-authorization-auditorSkill
- skill-authoring-toolkitSkill
- skill-evaluation-benchmarkSkill
- software-architectSkill
- technical-project-orchestratorSkill
- theme-design-systemSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Use for app illustrations, logos, icons, posters, printable artwork, social cards, responsive hero visuals and campaign graphics.
- Distinctive: line 14
Write a structured brief: hierarchy, concept and one visual focal point, tone, content, constraints, reference inspiration versus …
- Distinctive: line 24
QA checklist: test | tool/evidence | PASS / FAIL / NOT TESTED.
- Distinctive: line 32
No hidden trackers, external SVG dependencies, malicious scripts, unlicensed celebrity/brand claims, private user media uploads to public …
≈ “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
Visual asset design — creative direction to verifiable handoff
When to apply
Use for app illustrations, logos, icons, posters, printable artwork, social cards, responsive hero visuals and campaign graphics. Do not call for an unrelated API/backend change.
Specification before rendering
- Establish medium, audience, placement, physical dimensions or screen size, aspect ratio, target platforms, pixel density, bleed/safe margin, available source files and whether an editable source is required. Resolve typography language, font licensing, brand colors, accessibility and legal claims.
- Distinguish reference media, user-owned source, brand assets, AI-generated original work and externally licensed images. Record rights, attribution, permitted modifications, expiry/territory and any approvals; unclear rights are a blocker for publication.
- Write a structured brief: hierarchy, concept and one visual focal point, tone, content, constraints, reference inspiration versus prohibited copying, responsive crop variants and localized text. Preserve user-supplied identities/details when editing and disclose uncertainty.
- Choose output path based on available tools: (a) generate/edit a rendered image if an authorized image tool is available and requested, (b) create an inspectable SVG/HTML vector where source is requested, or (c) give a production-ready specification and clearly label it NOT RENDERED. Never claim an image file exists because a prompt was written.
- Specify exact sizes and units (px, mm, dpi as appropriate), intended exports (SVG/PNG/JPG/PDF/print CMYK subject to service capability), sharpness/contrast, RTL shaping, text safe areas and source format. Do not present a raster screenshot as an editable layered design.
- QA at final size: spelling and line breaks (especially Persian/Arabic), font fallback, copyright/trademarks, actual contrast, alignment, small-screen cropping, low-vision alternatives, transparency, bleed and printer profile. State which checks were visually inspected vs only specified.
- Iterate against measured acceptance criteria and record versioned filenames, licensed inputs, source file and exports. Do not modify/destroy originals without explicit permission.
Required deliverables
- Asset brief: intended use, audience, dimensions, copy, hierarchy, palette, reference constraints, localizations.
- Production matrix: artifact | editable source | export formats | dimensions | crop/bleed/safe zones | accessibility.
- Rights register: origin | owner/license | permitted use | credits/approval | status.
- QA checklist: test | tool/evidence | PASS / FAIL / NOT TESTED.
- Handoff decision: READY only for actually generated, rights-cleared and inspected assets; otherwise DRAFT / BLOCKED and exact next step; mark unavailable tool checks NOT RUN.
Local reference and web fallback
examples/visual-asset/ contains an authored, accessible SVG card and standard-library checks for viewBox/dimensions, descriptive text, references and rights manifest. These do not establish print color accuracy, legal rights or visual taste. In ChatGPT web, use available image tools when appropriate; if not, produce the brief and exact prompts/source code with the label NOT RENDERED. Do not pretend an AI image was created without an actual rendering tool.
Security, rights, and privacy
No hidden trackers, external SVG dependencies, malicious scripts, unlicensed celebrity/brand claims, private user media uploads to public stores or silent likeness modifications. Preserve original source material; document every changed element.
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.

