Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Design Iterate In Place

ASecurity

Use when iterating an existing design in place with A/B variants, an editable artboard, and anti-slop moodboards instead of generating from zero. Triggers on \"iterate design\", \"design variants\", \"artboard edit\", \"click to edit\", \"anti-slop\". Non-triggers: blank-page prototype generation (use cloud-design-prototyping), generic UI polish rules (use frontend-design-direction). Outcome: variant set against the existing system plus an edited winner ready to implement.

2 stars
0 votes
0 copies
0 views
Added 9/19/2026
ai-agentsgofrontend

Works with

cli

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add majinmagros/magros.ai-skills --skill design-iterate-in-place --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Design Iterate In Place?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Design Iterate In Place
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majinmagros-design-iterate-in-place/badge)](https://www.skillsdirectory.com/skills/majinmagros-design-iterate-in-place)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: design-iterate-in-place
description: "Use when iterating an existing design in place with A/B variants, an editable artboard, and anti-slop moodboards instead of generating from zero. Triggers on \"iterate design\", \"design variants\", \"artboard edit\", \"click to edit\", \"anti-slop\". Non-triggers: blank-page prototype generation (use cloud-design-prototyping), generic UI polish rules (use frontend-design-direction). Outcome: variant set against the existing system plus an edited winner ready to implement."
metadata:
  origin: ECC
---

# Design Iterate In Place

Iterate mid-process on a design that already exists: read its system,
produce variants, edit on an artboard, and kill AI slop per section.
Rule: respect the existing design system; this skill is mid-process,
not zero-to-one.

## When To Activate

- The user says iterate this design, variants A/B/C/D, artboard,
  click-to-edit, or fix this hero/landing.
- A design system or live page exists and must be respected.
- Output looks "AI-made" (neon glow overuse, repeated logo, 3+ text
  blocks, weak grid) and needs a human finish.
- A previous run regenerated from zero and broke brand consistency.

## Workflow

1. Read first: extract tokens, type scale, spacing, and components from
   the existing design or page. List what must not change.
2. Diagnose: name the competing elements (for example three dashed bands
   fighting, accent color on low-priority spots) in one short list.
3. Moodboard per section: collect 2-3 consistent references per section
   before generating. Famous skills all converge to "AI look" - the
   moodboard is the antidote.
4. Generate variants: produce A/B/C/D against the same brief (cleaner
   options first, one braver option last). Same content, different
   hierarchy and rhythm.
5. Artboard edit: open the winner on an editable artboard (properties
   panel, click-to-edit injects into the prompt). Edit in place, never
   restart from a blank canvas mid-round.
6. Human finish: fix grid and margins, redo type by hand where needed,
   add texture or glass sparingly, and run a side-by-side A/B before
   locking.
7. Hand off: ship the edited variant plus a delta list (what changed vs
   the original system) for implementation.

## Anti-Patterns

- Zero-to-one regeneration when a system already exists.
- Variant soup with no diagnosis of what is broken.
- One global moodboard for the whole page (slop returns per section).
- Editing screenshots by hand instead of via artboard -> prompt loop.
- Shipping without a side-by-side A/B against the original.

## Relations

- `cloud-design-prototyping`: external prototype flow; this skill is the
  in-place counterpart for existing designs.
- `frontend-design-direction`: product design judgment reused in diagnosis.
- `impeccable-design`: live multi-version editing behind the artboard step.

## Sources

- No external design-tool URL confirmed at write time - verify the native
  design skill and artboard protocol of your current harness before use.
- Variants A/B/C/D, Figma-style editable artboard (properties panel,
  click-to-edit into prompt), mid-process-not-zero-to-one rule, and the

## Exemplo

```text
Landing com hero poluído (3 faixas tracejadas + glow neon + logo repetido)
Diagnóstico: 3 elementos brigando → moodboard por seção (não global)
Variantes A/B/C/D no mesmo brief → vencedora no artboard (click-to-edit)
Fecha: A/B lado a lado vs original + delta list p/ implementação
```

## Swarm de variantes lado-a-lado vs dividido (leva YouTube rodada 10)

Dois modos: DIVIDIR (coordenador + 2 workers, cada um numa direcao) vs LADO-A-LADO (mesma tarefa, 2-6 variantes com layout diferente ou estilo diferente). Cap de frames para economizar tokens. Passe o pre-check em swarm-readiness-gate antes de subir o enxame.

Attribution

majinmagrosmajinmagros
View sourceMore from majinmagros →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3331 votes

catchup

Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.

611 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →