メニュー
When a ticket must go to a human before any reply is sent, and how the first reply should read. Use when drafting a customer reply or setting needs_human_review.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/Kiln-AI/Kiln --skill escalation-and-reply-toneコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Look at Kiln's UI in a real browser, and run its end-to-end tests. Use when checking UI you are changing, taking a screenshot of the app, driving the app with playwright-cli, starting the dev sandbox with playwright_server.sh, running or debugging `npm run tests:e2e`, or working with the seeded fixture project that gives the app's screens data to render.
Multi-agent, browser-driven QA pass over a branch or PR — scope it, plan it, fan out one subagent per testing area in its own isolated sandbox, compile a severity-ranked markdown report. Use for an "E2E test", "QA pass", "manual test", or "bug bash" request — broader than driving the UI to check one fix (see `playwright`) or reviewing a diff (see `code-review`).
Add new AI models to Kiln's ml_model_list.py and produce a Discord announcement. Use when the user wants to add, integrate, or register a new LLM model (e.g. Claude, GPT, DeepSeek, Gemini, Kimi, Qwen, Grok) into the Kiln model list, mentions adding a model to ml_model_list.py, asks to discover/find new models that are available but not yet in Kiln, or wants to add a net-new AI provider to Kiln.
SOC 職業分類に基づく
SKILL.md を表示中
| name | escalation-and-reply-tone |
| description | When a ticket must go to a human before any reply is sent, and how the first reply should read. Use when drafting a customer reply or setting needs_human_review. |
Use this skill when drafting the first reply to a customer, or when deciding whether a ticket needs a human before anything is sent.
Set needs_human_review to true, and do not send an automated reply, when the ticket involves any of:
Urgency alone is not escalation. A production outage with a known cause and a workaround is urgent and still does not need human review.
Open with what you understood, in the customer's own terms, before saying what you will do. Never open with an apology template.
Say what happens next and who does it. "I have routed this to our billing team, who will reply today" beats "your ticket has been escalated".
Do not promise a timeline the team has not committed to. If you do not know when something will be fixed, say what you will tell them and when.
Match the customer's register. A one-line question gets a one-line answer; a detailed report gets a reply that addresses each part in the order they raised it.
