user-context

作成者: openai

Data Analyticsプラグインの永続的なソースルーティング設定、オンボーディングロジック、セットアップ進捗、およびセマンティックレイヤーレジストリを読み込むか管理します。

npx skills add https://github.com/openai/role-specific-plugins --skill user-context

User Context

User Context stores the product and design references a designer uses often, so future Product Design work starts from the right sources.

Use this skill when the user asks to:

  • set up Product Design
  • get started with Product Design
  • onboard with Product Design
  • save product or design sources
  • see what Product Design remembers
  • update saved product or design context
  • remember a Product Design preference
  • setup my plugin

Critical Overrides

  • Refer to the Plugin router $index before proceeding.
  • Follow $critical-overrides.
  • Before offering onboarding or saving context for future conversations, confirm that local shell access is available, $CODEX_HOME can be resolved, and $CODEX_HOME/state/plugins/product-design/ exists and is writable or can be created in a writable parent directory. If user-context.md already exists, confirm that it is writable too.
  • If any check cannot be completed or fails, persistent context is unavailable. Do not offer saved-context onboarding or claim that new context was saved for future conversations. If the user asks to save something, explain that it can be used in the current conversation but not saved for future conversations.

Saved User Context

If user-context.md exists, use it by default.

Use saved product URLs, Figma files, screenshots, reference images, codebase paths, Storybook, tokens, design systems, brand assets, component refs, browser preferences, and share targets to ground Product Design work.

Ideation, prototypes, audits, clones, and critiques should match the saved product context unless the user asks for something different.

When a workflow needs visual grounding, attach or include relevant saved screenshots, reference images, tokens, design language, and component references in ImageGen, ideation, prototype, audit, and critique work.

State File

Saved context lives here:

$CODEX_HOME/state/plugins/product-design/user-context.md

Saved screenshots and reference images live next to it:

$CODEX_HOME/state/plugins/product-design/assets/

If the file does not exist, continue normally unless the user asks to set up Product Design, save context, or the current task is blocked by missing product/design context.

Preflight

When any Product Design workflow needs saved context, run:

python3 scripts/user_context_preflight.py

Use the returned saved entries as the starting context for the task.

If the script reports that no saved context exists, continue from the current user prompt unless setup context is required.

Do not browse, open, or inspect every saved reference during preflight. Inspect only the saved references needed for the current task.

Setup

Use references/onboarding.md when the user asks to set up Product Design, asks what Product Design can remember, asks what Product Design knows about their product, or provides product/design references to save.

For setup-only requests, explain what Product Design can remember and ask for useful sources.

Adjust the context-gathering request to match the user's request. First-time setup differs from updating existing context.

Do not inspect the workspace, install dependencies, scaffold a prototype, generate images, run audits, or start implementation during setup.

After the user provides references to save, run:

python3 scripts/init_user_context.py

Then add the references to the created user-context.md.

Save

Save useful, durable Product Design context:

  • Product URLs
  • Figma files
  • Screenshots and reference images
  • Codebase paths
  • Storybook and component docs
  • Design tokens and theme sources
  • Brand, logo, icon, illustration, image, and asset sources
  • Preferred browser, capture tools, and share targets
  • Team conventions that make future Product Design work more accurate

When the user provides screenshots or reference images to save, copy them into assets/ next to user-context.md and link them from the saved entry.

Give each saved image a clear, descriptive filename that says what the image shows. Use names future Product Design runs can understand without opening the file.

Good image names:

assets/chatgpt-settings-modal-dark-mode.png
assets/payment-sheet-mobile-error-state.png
assets/product-dashboard-sidebar-navigation.png
assets/storybook-primary-button-states.png
assets/brand-logo-lockup-purple-gradient.png
assets/onboarding-flow-welcome-step.png
assets/checkout-confirmation-screen.png
assets/account-menu-open-state.png

Do not save secrets, credentials, API keys, private tokens, copied customer data, or anything that should not persist.

Use this structure:

# {Category}

- Description: {what this category is and when future Product Design runs should use it}

## Saved Links And Context

{Saved reference or fact}
- Date Added: YYYY-MM-DD.
- File: assets/{clear-descriptive-name}.png
- Useful Context: {what this reference represents}
- Future Use: {how future Product Design work should use it}

Include File: only when the saved entry has a local image file.

When a category has no saved references yet, use exactly:

status: not provided

Keep saved context curated. Prefer a few high-value references over a dump of every possible URL or file.

Read

  • Do not treat status: not provided as a fact.
  • Read through scripts/user_context_preflight.py when local shell access is available.
  • Use saved context as default grounding, then inspect only what the current task needs.

openaiのその他のスキル

notion-research-documentation
openai
Notionのコンテンツを調査し、構造化されたブリーフ、レポート、または引用付きの比較にまとめます。対象クエリを使用してNotionページを検索・取得し、テーマごとに調査結果を整理し、インラインのソース引用と参考文献セクションを追加します。範囲とユーザーの目的に応じて、4つの出力形式(クイックブリーフ、調査サマリー、比較、包括的レポート)から選択します。組み込みテンプレートを使用してNotionページを作成・更新し、新しい情報が到着するたびにソースを直接リンクし、変更を追跡します...
official
rcsb-pdb-skill
openai
コアメタデータ、Search APIクエリ、FASTAダウンロードのためのコンパクトなRCSB PDBリクエストを送信します。ユーザーが簡潔なRCSBサマリーを希望する場合に使用し、生のJSONや…を保存します。
official
pdf
openai
PDFの読み取り、作成、検証(視覚的レンダリングおよびプログラムによる生成を含む)。Poppler(pdftoppm)を使用して、納品前にレイアウト、スペーシング、タイポグラフィを視覚的に検査するためにPDFページをPNGにレンダリング。reportlabを使用してプログラム的にPDFを生成し、信頼性の高いフォーマットを実現。pdfplumberまたはpypdfを使用してテキストとメタデータを抽出。品質基準を遵守:テキストのクリッピングなし、要素の重なりなし、テーブルの破損なし、レンダリングアーティファクトなし。ASCIIハイフンのみ、人間が読める引用を使用。
official
test-coverage-improver
openai
Improve test coverage in the OpenAI Agents JS monorepo: run `pnpm test:coverage`, inspect coverage artifacts, identify low-coverage files and branches, propose…
official
playwright
openai
ターミナル駆動のブラウザ自動化で、要素スナップショットとインタラクティブなUIワークフローを備えています。playwright-cliラッパースクリプト(npxが必要)を介して動作し、ヘッドレスモードとヘッドモードの両方をサポートし、ビジュアルデバッグが可能です。コアワークフローは、ページを開き、安定した要素参照のためにスナップショットを取得し、参照を使用して操作し、ナビゲーションやDOM変更後に再スナップショットを取得します。フォーム入力、クリック、タイピング、マルチタブ管理、スクリーンショット/PDFキャプチャ、フローデバッグ用のトレース記録を含みます。要素参照(例:e3、e15)...
official
ukb-topmed-phewas-skill
openai
単一バリアントのコンパクトなUKB-TOPMed PheWASサマリーを取得します。rsID、GRCh37、またはGRCh38の入力を受け付け、必要なGRCh38クエリに解決します。以下の場合に使用します…
official
code-review-context
openai
モデル可視コンテキスト
official
signing-entitlements
openai
macOSアプリの署名、権限、ハード化されたランタイム、Gatekeeperの問題を検査します。コード署名の失敗や権限の欠落などの診断を求められた際に使用します。
official