本文へ移動
cccskills
無料GitHub で公開

grill-me

Interview the user about any topic one question at a time, using the grilling engine with a question set generated from the prompt. Use for "grill me on X" outside of planning. Pass --rounds for batched rounds.

日本語の概要は準備中です。原文の説明を表示しています。

indigoai-us/hq-coreスター 90(GitHub 90 + cccskills 0)

実行スコア 未検証

インストール方法を見る

含まれるファイル(2)

  • SKILL.md2.9 KB
  • agents/openai.yaml115 B

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

/grill-me

Thin entry to the grilling engine (.claude/skills/grilling/SKILL.md). This skill builds the question set; the engine asks it.

Arguments

  • <topic>: what to grill the user on. If empty, ask for it in one AskUserQuestion call.
  • --rounds: pass mode: rounds to the engine. Without it, pass mode: one-at-a-time. Never switch modes on your own.

Procedure

  1. Anchor. Resolve the company: explicit mention in the topic, then the cwd's company folder, then the repo's owner in companies/manifest.yaml, then the session's company_slug. When a company resolves, read its policies folder. If none resolves, run unscoped and read no company files. Set project_dir when the topic or cwd names a project folder of that company; otherwise leave it unset.
  2. Research facts. Read what the topic points to (board, repo, knowledge) so known answers are not asked. Record settled answers as known_facts.
  3. Generate the question set. Write workspace/tmp/grill-me/{date}-{topic-slug}.md in the format in .claude/skills/_shared/questions/FORMAT.md:
    • 4 to 12 questions. At least 4 must be kind: decision.
    • Cover tiers in order: strategic (goal, scope, audience), architecture (approach, boundaries), quality (risks, success measure).
    • Each decision has 2 to 4 short options and a recommended option.
    • Use depends_on where one answer changes the next question. Use kind: fact for anything you can look up.
  4. Validate. bash .claude/skills/grilling/frontier.sh count <set> --known <ids>. On exit 2, fix the set and validate again.
  5. Open a journal when project_dir is set: bash .claude/skills/_shared/journal.sh open grill-me <project_dir>.
  6. Run the engine with question_set, known_facts, and mode. Follow its procedure exactly.
  7. Summarize. Write a short summary: topic, each decision and its answer, assumed answers, open items, and the count object.
    • Always write it to the journal: journal.sh append <project_dir> decisions "summary: ..." when a project journal is open, otherwise a /journal session entry.
    • When project_dir is set, also write it to {project_dir}/grill-{YYYY-MM-DD}.md, then run journal.sh close <project_dir> "<one-line summary>".
  8. Report in chat: the decisions in plain words and where the summary was saved. When the topic is a project, suggest /brainstorm or /prd as the next step.

Rules

  • One AskUserQuestion call per question unless --rounds was passed.
  • Never answer your own decision questions, except under CLAUDE_HEADLESS=1 as the engine defines.
  • Do not read another company's files or credentials.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

コメント

ログインなしで書き込めます。使ってみた感想、質問、動かなかった環境などを気軽にどうぞ。

まだコメントはありません。最初のコメントを書いてみませんか。

送信するとすぐに公開されます。メールアドレスや電話番号などの個人情報は書かないでください。URL は 2 つまでです。

0 / 1000 文字

このフォームは reCAPTCHA で保護されており、Google のプライバシーポリシーと利用規約が適用されます。

同じリポジトリのスキル

accept

無料

Accept HQ membership invites from a magic link or raw token.

日本語の概要は準備中です。原文の説明を表示しています。

indigoai-us/hq-core902026年10月26日 更新

adr

無料

Capture qualifying architecture decisions as ADRs in the right HQ or repo location.

日本語の概要は準備中です。原文の説明を表示しています。

indigoai-us/hq-core902026年10月26日 更新

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

日本語の概要は準備中です。原文の説明を表示しています。

indigoai-us/hq-core902026年10月26日 更新

architect

無料

Find shallow modules and propose deepening refactors, ranked with file refs; never edits code. Use when code is hard to change or planning a refactor.

日本語の概要は準備中です。原文の説明を表示しています。

indigoai-us/hq-core902026年10月26日 更新

Generate dark ASCII banner graphics for HQ and social launch assets.

日本語の概要は準備中です。原文の説明を表示しています。

indigoai-us/hq-core902026年10月26日 更新

Compare approaches to best-in-class prior art, merge existing work, verify cruft removal, and recommend the most durable design before PRD work.

日本語の概要は準備中です。原文の説明を表示しています。

indigoai-us/hq-core902026年10月26日 更新

indigoai-us のスキルをすべて見る

このスキルの問題を報告する