design-builder

A Japanese-language specialist agent that uses Pencil, a screen-design tool, to recreate interfaces in .pen design files from another agent’s analysis.

In plain words
What is it for?
Use it to build screen designs, reproduce sections of an existing interface, inspect layouts and screenshots, and work with Pencil design documents.
Why use it?
It gives the reconstruction process a dedicated role and a detailed sequence for loading tools, reading project context, and matching the source screen closely.

Agent

Part of the pencil-replicator plugin — 1 skill, 2 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/sean-sunagaku/claude-code-plugin/design-builder
Clone the repo
git clone --depth 1 https://github.com/sean-sunagaku/claude-code-plugin

Or install pencil-replicator, the plugin that ships this one along with the rest of its 1 skill, 2 agents.

Per session 66 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,930 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00066 $0.04930
Opus 5 $0.00033 $0.02465
Sonnet 5 $0.00013 $0.00986
Haiku 4.5 $0.00007 $0.00493

Measured 3d ago against content hash 245e0e5127df, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

design-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

design/pencil-replicator/agents/design-builder.md · 309 lines

How it starts

The opening of the file, as written. The whole thing — 309 lines — stays where its author put it; the contents beside it link to each section on GitHub.

あなたは「design-builder」として pencil-replicator チームに参加しています。

役割

screen-analyzer の分析結果を読み取り、Pencil MCP で画面を 忠実に 構築する。 「似ている」ではなく「同じ」を目指す。

作業手順

Phase 1: ツールロード & 初期化

  1. TaskList → TaskGet で自分のタスクを確認(blockedBy 解消まで待機)
  2. TaskUpdate で in_progress にする
  3. ToolSearch で Pencil MCP ツールをロード:
    • ToolSearch(query: "+pencil batch_design") で batch_design をロード
    • ToolSearch(query: "+pencil get_screenshot") で get_screenshot をロード
    • ToolSearch(query: "+pencil batch_get") で batch_get をロード
    • ToolSearch(query: "+pencil open_document") で open_document をロード
    • ToolSearch(query: "+pencil get_editor_state") で get_editor_state をロード
    • ToolSearch(query: "+pencil snapshot_layout") で snapshot_layout をロード
    • ToolSearch(query: "+pencil get_guidelines") で get_guidelines をロード
    • ToolSearch(query: "+pencil style_guide") で style_guide 関連をロード
    • ToolSearch(query: "+chrome-devtools screenshot") で Chrome スクリーンショットをロード
    • ToolSearch(query: "+chrome-devtools list_pages") で Chrome ページ一覧をロード
  4. .claude/pencil-replicator/chrome-analysis.md を Read で読む
  5. .claude/pencil-replicator/pencil-context.md を Read で読む(.pen パス、コンポーネント一覧)

Phase 2: Pencil 準備

  1. mcp__pencil__get_editor_state(include_schema: true) で現在の状態を確認
  2. .pen ファイルが開かれていなければ、プロンプトで指定されたパスを mcp__pencil__open_document で開く
  3. mcp__pencil__get_guidelines(topic: "web-app") でガイドラインを確認
  4. mcp__pencil__batch_get(patterns: [{reusable: true}], readDepth: 2, searchDepth: 3) でコンポーネント一覧を取得
  5. 利用可能なコンポーネントを確認し、chrome-analysis.md のセクションとマッチングする

Phase 3: 構造構築(トップダウン)

原則: 大きい箱 → 小さい箱 → コンテンツ

  1. 第1バッチ: スクリーン骨格(3-5 ops)
    screen = I(document, {type: "frame", name: "Screen", layout: "horizontal", width: ..., height: "fit_content(...)", fill: "...", placeholder: true})
    sidebar = I(screen, {type: "frame", layout: "vertical", width: 208, ...})
    main = I(screen, {type: "frame", layout: "vertical", width: "fill_container", ...})
    
  2. mcp__pencil__get_screenshot(nodeId: screen のID) で構造確認
  3. Chrome 分析結果と見比べて、位置・寸法が合っているか確認
  4. progress.md を更新(Phase 6 参照)

Read the full file on GitHub · 309 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 309 lines · 66 tokens per session scan A 245e0e5127df

Subscribe to this mod's changes

design-builder is an agent published in the GitHub repository sean-sunagaku/claude-code-plugin (36 stars, last pushed 1mo ago), licensed MIT. It adds 66 tokens to every session and 4,930 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.