본문으로 건너뛰기
cccskills
무료GitHub에 공개

iterative-design-assistant

Track design iteration history and apply consistent styling decisions across revision cycles. Use when handling follow-up change requests that reference previous design decisions (e.g., '前回も色で良いんだけど', 'same style as before', 'like we discussed').

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

연동 도구PowerPoint
설치 방법 보기

포함된 파일(5)

  • SKILL.md6.5 KB
  • references/design-decision-methodology.md8.0 KB
  • scripts/design_log.py19.1 KB
  • scripts/tests/conftest.py239 B
  • scripts/tests/test_design_log.py17.6 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Iterative Design Assistant

Overview

This skill maintains a session-local design decision log that tracks styling choices, revision history, and contextual relationships across multiple document or presentation revision cycles. It enables Claude to understand references to previous changes and apply consistent design decisions without requiring users to repeat context.

When to Use

  • User references a previous design decision ("前回も色で良いんだけど", "like last time", "same as before")
  • Multiple revision cycles on the same document/presentation
  • Need to track which design elements were changed and why
  • Applying consistent styling across related documents
  • Reviewing design history to understand evolution of a document
  • User asks to "undo" or "revert" to a previous design state

Prerequisites

  • Python 3.9+
  • No API keys required
  • Standard library only (json, datetime, pathlib)

Workflow

Step 1: Initialize Design Session

When starting work on a document/presentation, initialize a design session log. The log file is created in the current working directory.

python3 scripts/design_log.py init \
  --document "presentation.pptx" \
  --session-name "Q4 Sales Deck Revisions"

This creates a .design-log.json file that tracks all design decisions for the session.

Step 2: Record Design Decisions

Each time a design change is made, record it in the log with context about what was changed and why.

python3 scripts/design_log.py record \
  --category "color" \
  --element "header-background" \
  --old-value "#FFFFFF" \
  --new-value "#003366" \
  --reason "User requested corporate blue for headers" \
  --reference "slide-3"

Categories include:

  • color - Color changes (hex values, color names)
  • typography - Font family, size, weight changes
  • layout - Positioning, spacing, alignment changes
  • content - Text, image, or data changes
  • style - Border, shadow, effect changes

Step 3: Query Previous Decisions

When user references a previous decision, query the log to retrieve context.

python3 scripts/design_log.py query \
  --category "color" \
  --limit 5

Or search by keyword:

python3 scripts/design_log.py search \
  --keyword "header" \
  --limit 10

Step 4: Apply Consistent Styling

When applying a previous decision to new elements, reference the decision ID.

python3 scripts/design_log.py apply \
  --decision-id "dec_001" \
  --target-element "slide-7-title"

This records that the same styling was applied to a new element, maintaining traceability.

Step 5: Review Design History

Generate a summary of all design decisions for documentation or review.

python3 scripts/design_log.py history \
  --format markdown \
  --output design-history.md

Step 6: Handle Contextual References

When user says something like "前回も色で良いんだけど" (the color from last time is fine), use the context resolution workflow:

  1. Query recent color-related decisions
  2. Identify the most relevant previous decision
  3. Apply the same value to the current context
  4. Record the new application with reference to the original decision
# Find recent color decisions
python3 scripts/design_log.py query --category color --limit 3

# Apply the identified decision
python3 scripts/design_log.py apply \
  --decision-id "dec_003" \
  --target-element "current-element" \
  --context "User confirmed previous color choice"

Output Format

Design Log JSON Schema

{
  "schema_version": "1.0",
  "session": {
    "id": "session_uuid",
    "name": "Q4 Sales Deck Revisions",
    "document": "presentation.pptx",
    "created_at": "2024-01-15T10:30:00Z",
    "updated_at": "2024-01-15T14:45:00Z"
  },
  "decisions": [
    {
      "id": "dec_001",
      "timestamp": "2024-01-15T10:35:00Z",
      "category": "color",
      "element": "header-background",
      "old_value": "#FFFFFF",
      "new_value": "#003366",
      "reason": "User requested corporate blue for headers",
      "reference": "slide-3",
      "applied_to": ["slide-3-header", "slide-7-title"],
      "superseded_by": null
    }
  ],
  "design_tokens": {
    "colors": {
      "primary": "#003366",
      "secondary": "#F0F0F0"
    },
    "typography": {
      "heading-font": "Arial Bold",
      "body-font": "Arial"
    }
  }
}

Markdown History Report

# Design History: Q4 Sales Deck Revisions

**Document**: presentation.pptx
**Session Started**: 2024-01-15 10:30

## Timeline

### 2024-01-15 10:35 - Color Change
- **Element**: header-background
- **Change**: #FFFFFF → #003366
- **Reason**: User requested corporate blue for headers
- **Applied to**: slide-3-header, slide-7-title

### 2024-01-15 11:20 - Typography Change
- **Element**: body-text
- **Change**: 12pt → 14pt
- **Reason**: Improved readability for presentation room

## Design Tokens

| Category | Token | Value |
|----------|-------|-------|
| Color | Primary | #003366 |
| Color | Secondary | #F0F0F0 |
| Typography | Heading Font | Arial Bold |
| Typography | Body Font | Arial |

Resources

  • scripts/design_log.py -- CLI tool for managing design decision logs
  • references/design-decision-methodology.md -- Best practices for tracking and applying design decisions

Key Principles

  1. Context Preservation -- Every design decision includes the reason and context, enabling accurate interpretation of future references
  2. Bidirectional Traceability -- Track which decisions were applied to which elements, and which elements derive from which decisions
  3. Incremental History -- Build up a design token vocabulary as decisions accumulate, enabling consistent future styling
  4. Natural Language Resolution -- Support vague references ("like before", "前回と同じ") by matching against recent relevant decisions
  5. Session Isolation -- Each document/session has its own log, preventing cross-contamination while allowing explicit cross-referencing

Integration with Other Skills

This skill complements:

  • fujisoft-presentation-creator -- Track design decisions made during presentation creation
  • presentation-reviewer -- Reference design history when reviewing presentations
  • markdown-to-pdf -- Apply consistent styling tokens to PDF generation

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Track and update action item status from natural language updates like 'Seanのメールには返信しておいた' or 'Lu対応予定'. Integrates with daily-comms-ops workflow. Use when managing action items across communication channels.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

takusaotome/claude-skills-library102026년 10월 5일 업데이트

AI活用提案の専門コンサルタントスキル。LLM・AIエージェントの活用事例知識を基に、業界別、部門別、シーン別に最適なAI導入提案を行います。システム化、業務改善、効率化を得意とし、具体的な活用事例、期待効果、導入ステップを含む実践的な提案を提供します。Use when proposing AI/LLM adoption strategies for specific industries, departments, or business scenarios.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

takusaotome/claude-skills-library102026년 10월 5일 업데이트

Generate AI-powered BPO service proposals for Japanese companies in the US market. Use when creating outsourcing proposals with AI service menus, ROI estimation, implementation roadmaps, and bilingual (JA/EN) proposal documents for 在米日系企業向けAI実装.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

takusaotome/claude-skills-library102026년 10월 5일 업데이트

AI(LLM)が生成した日本語テキストの「AI臭」を検出・診断し、人間らしい文章にリライトするスキル。 Use when: 「AIっぽい文章を直して」「人間らしくリライトして」「AI臭を消して」「この文章をもっと自然にして」 「テキストのAI感を減らして」「機械っぽさを取りたい」 "make this sound more human", "remove AI tone", "humanize this text", "detect if this is AI-generated". 6つのAI特有パターン(視覚的マーカー、単調なリズム、マニュアル的構成、非コミット姿勢、抽象語の濫用、 定型メタファー)を正規表現ベースで検出し、0-100のAI臭スコアを算出。3つの人間化技法 (バランスを崩す・客観を崩す・論理を崩す)でリライトを実行する。 Note: 検出スクリプトは日本語テキスト専用。英語テキストの場合はClaude自身がreferences/を参照して分析・リライトする。

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

takusaotome/claude-skills-library102026년 10월 5일 업데이트

Generate audit-ready internal control design documents from As-Is business process inventories. Produces control IDs, assertion mappings, procedures, SoD analysis, KPIs, materiality thresholds, and implementation roadmaps. Use when building internal controls for new audit engagements, SOX/J-SOX compliance, or process improvement initiatives.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

takusaotome/claude-skills-library102026년 10월 5일 업데이트

Review audit-related documents (control design documents, bottleneck analyses, requirements definitions, etc.) for quality, scoring them 0-100 with a severity-rated findings list. Use when reviewing audit documents, checking control design quality, or verifying cross-document consistency. Supports documents governed by US GAAP, IFRS, or J-GAAP.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

takusaotome/claude-skills-library102026년 10월 5일 업데이트

takusaotome의 스킬 모두 보기

이 스킬의 문제 신고