Install
$ agentstack add skill-ceilf6-frontagent-frontend-design ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README — it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming — see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps — measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Frontend Design
Use this skill to produce frontend work that feels intentionally designed rather than generic. The goal is not maximalism by default. The goal is a clear visual point of view that matches the product context and still ships as maintainable code.
Trigger
- New frontend surfaces where look and feel matter:
- landing pages
- dashboards
- product pages
- hero sections
- marketing sites
- polished application views
- Requests to beautify, restyle, refine, or modernize an existing UI
- Cases where the code must be working and production-usable, not just a mockup
Do Not Use
- Pure bug fixes, refactors, or backend tasks
- Review or audit requests
- Small code changes where the user did not ask for visual improvement
- Existing design-system work where the user only wants strict conformance and no meaningful visual interpretation
Workflow
- If the request is visually underspecified, ask only the next 1-3 questions needed to clarify audience, tone, and acceptance criteria. Use
requirement-interviewerwhen the ambiguity is material. - Read
references/design-direction.mdto choose one clear visual direction before coding. - Read
references/aesthetics-guidelines.mdfor typography, color, motion, composition, and background treatment. - Read
references/integration-guardrails.mdbefore modifying an existing codebase, design system, or shared component surface. - For a new page or major redesign, use
assets/templates/visual-brief.mdto form a short visual brief before implementation. - Implement working code that matches the chosen direction and stays within project constraints.
- Read
references/delivery-checks.mdbefore handing off.
Output Contract
- State the chosen visual direction in one short sentence when it meaningfully affects implementation.
- Produce working frontend code, not mood-board prose.
- Keep the implementation cohesive:
- typography should support the direction
- color should feel intentional
- motion should be meaningful, not noisy
- layout should feel composed rather than assembled from defaults
- When editing an existing product, preserve the established system unless the user explicitly wants a stronger redesign.
Guardrails
- Project design specification constraints, protected files, project stack rules, and user requirements remain hard boundaries.
- Respect the existing design system first. Only deviate when the request explicitly calls for redesign or the current surface clearly lacks a defined visual direction.
- Prefer minimal diffs for scoped tasks. Do not turn a local UI tweak into a whole-app restyle.
- Do not add new dependencies, font loaders, or animation libraries unless the task clearly justifies them.
- Avoid generic AI aesthetics:
- default-looking font stacks for new standalone surfaces
- timid palettes
- predictable card grids with no visual idea
- purple-gradient-on-white defaults
- Do not optimize for ornament over clarity. If visual flair hurts usability, accessibility, or maintainability, scale it back.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ceilf6
- Source: ceilf6/FrontAgent
- License: MIT
- Homepage: https://marketplace.visualstudio.com/items?itemName=ceilf6.frontagent
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.