.cursor/skills/terse-japanese-thinking/SKILL.md
Use impolite, maximally terse Japanese (タメ口/ぞんざい語) in thinking blocks and subagent prompts to conserve tokens. User-facing output remains in the user's language. Use proactively on every response that involves internal reasoning or subagent delegation.
npx skillsauth add Intersubjective/tentura terse-japanese-thinkingInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
All <thinking> block content and all Task tool prompt values MUST be in raw, blunt Japanese. User-facing text stays in the user's language.
Bad (wasteful):
Let me think about this. The user wants to add a new endpoint.
I should first check the existing router configuration to understand
the current patterns being used. Then I'll need to create the handler
function and register the route. Let me also consider error handling...
Good (terse Japanese):
新エンドポイント追加。
- router設定確認する
- handler作る
- ルート登録
- エラー処理要る
先にrouter.dart読む
Bad (polite Japanese — token waste):
ユーザーさんが新しいエンドポイントを追加したいとおっしゃっています。
まず、既存のルーター設定を確認させていただきたいと思います。
Good:
endpoint追加。router見る。
Bad (verbose English prompt to Task tool):
Please search the codebase for all files that define a repository class
extending RemoteRepository. I need to understand how they handle error
cases and what patterns they follow for retry logic. Return a summary
of each repository found with its file path and key methods.
Good (terse Japanese prompt to Task tool):
RemoteRepository継承してるrepositoryクラス全部探せ。
- ファイルパス
- エラー処理パターン
- retry有無
まとめて返せ。
Bad (polite Japanese prompt — token waste):
リポジトリクラスを検索していただけますか。
RemoteRepositoryを継承しているものを探していただきたいです。
Good:
RemoteRepository継承クラス全探索。パス・エラー処理・retry有無返せ。
prompt field): Japanese.description field: English (short, 3-5 words as required by tool).prompt): always English (tool names, file paths, commands).development
Applies Hasura metadata changes (track tables, relationships, permissions) by editing hasura/metadata.json and running scripts/hasura_apply_metadata.sh against local dev compose. Use when tracking a new Postgres table in Hasura, adding relationships or permissions, or when the user asks to update Hasura metadata.
tools
Implements a Figma selection or node URL as production-quality Flutter UI for this monorepo (packages/client), using MCP design context and screenshot verification. Use when the user mentions Figma, a frame, node, component, screen, MCP, or "implement this design"; when a Figma selection is active; or when a Figma frame/layer link is provided for Flutter UI work.
testing
Create, edit, improve, or audit AgentSkills. Use when creating a new skill from scratch or when asked to improve, review, audit, tidy up, or clean up an existing skill or SKILL.md file. Also use when editing or restructuring a skill directory (moving files to references/ or scripts/, removing stale content, validating against the AgentSkills spec). Triggers on phrases like "create a skill", "author a skill", "tidy up a skill", "improve this skill", "review the skill", "clean up the skill", "audit the skill".
testing
Host security hardening and risk-tolerance configuration for OpenClaw deployments. Use when a user asks for security audits, firewall/SSH/update hardening, risk posture, exposure review, OpenClaw cron scheduling for periodic checks, or version status checks on a machine running OpenClaw (laptop, workstation, Pi, VPS).