kaji: Skill for Claude Code

.claude/skills/grill-me/SKILL.md

grill-me is a skill for Claude Code, Codex from apokamo/kaji. It costs 95 tokens per session (2,121 once invoked), scanned A, original, Apache-2.0.

A guided interview for resolving important, still-undecided choices in an issue before its workflow starts. It asks one question at a time and records confirmed decisions and their origins.

In plain words
What is it for?
It is for reviewing a specific issue, resolving its decision branches with a human, and writing the agreed decisions back to that issue.
Why use it?
It prevents major decisions from remaining unclear when work begins and avoids asking people about facts that can be found in the code or documentation.

Skill for Claude CodeCodex

Written for Claude Code and Codex: $ARGUMENTS substitution, but also agents/openai.yaml present.

This is apokamo/kaji's own configuration. It tells Claude Code and Codex how to work on kaji itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything kaji configures →

Reuse

Borrowing it

Nothing to install: this file belongs to apokamo/kaji. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/apokamo/kaji/main/.claude/skills/grill-me/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/apokamo/kaji

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for grill-me

README.md
[![agentmods](https://agentmods.dev/badge/skills/apokamo/kaji/grill-me/github.svg)](https://agentmods.dev/skills/apokamo/kaji/grill-me)
Your own site
<a href="https://agentmods.dev/skills/apokamo/kaji/grill-me"><img src="https://agentmods.dev/badge/skills/apokamo/kaji/grill-me/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for grill-me

Your own site · 80×15
<a href="https://agentmods.dev/skills/apokamo/kaji/grill-me"><img src="https://agentmods.dev/badge/skills/apokamo/kaji/grill-me.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 95 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,121 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 5.1 $0.00095 $0.02121
Opus 5 $0.00048 $0.01060
Sonnet 5 $0.00019 $0.00424
Haiku 4.5 $0.00010 $0.00212

Measured 10d ago against content hash 0477dfa0c2f0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

grill-me 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 10d 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.

.claude/skills/grill-me/SKILL.md · 151 lines

How it starts

The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Grill Me

重要判断を workflow の有人区間で前倒しして確定する。質問数を減らすことではなく、 shared understanding に達するまで未決分岐を残さないことを優先する。ただし、調査で 判明する事実や後段で安く直せる詳細を人間へ質問しない。

位置と責務境界

issue-create → (/grill-me: 任意・明示起動・有人) → issue-review-ready → issue-start → …
  • .kaji/wf/**/*.yaml の step には追加しない。headless workflow に同期対話を持ち込まない。
  • issue-review-ready を代替しない。interview 後も独立した readiness gate を通す。
  • 明示起動専用とする。対応 runtime が agents/openai.yaml を読む場合は policy.allow_implicit_invocation: false に従い、それ以外でもこの使用条件を守る。
  • 判断軸、可逆性、停止条件、provenance の正本は _shared/critical-decision-checklist.md とする。

入力

$ARGUMENTS = <issue_id>

人間が対話に参加できる状態で /grill-me <issue_id> と明示起動する。Issue ID がない、 Issue を取得できない、または回答者が不在なら、Issue を変更せず停止する。

Interview の 5 不変条件

  1. 全 aspect の未決分岐を shared understanding に達するまで追跡する。
  2. 決定間の依存順に decision tree の branch を 1 つずつ解決する。
  3. 1 回に提示する質問は必ず 1 問とし、回答を受けるまで次を提示しない。
  4. コード、docs、既存 Issue / PR で回答できる事項は先に調査し、人間へ聞かない。
  5. 各質問に推奨回答と短い理由を付ける。

加えて、人間の確認なしに決定事項を確定または Issue へ書き戻さない。

relentless は質問を省略しないという意味であり、攻撃的に問い詰めることではない。 前置きを短くし、判断コストを下げる協働的な質問にする。

実行手順

Step 1: Issue と正本を読む

  1. Issue 本文と全コメントを読む。
  2. _shared/critical-decision-checklist.md を読む。
  3. Issue が参照するコード、docs、既存 Issue / PR、外部仕様を調査する。
  4. 事実、人間が既に決定した事項、AI の仮定、真に未決の事項を分離する。

source of truth 同士が矛盾する場合は、都合のよい方を選ばず未決分岐に戻す。

Step 2: 要否門番を通す

次のいずれかを含みうる場合は interview を続ける。

  • ユーザー価値、主要ユースケース、画面・操作構造
  • 公開 CLI / API、永続化形式、migration、rollback
  • source of truth の指定または競合
  • 既存利用者、外部連携、運用、権限、監視への影響
  • 非互換変更、quality gate の省略、曖昧なスコープ境界

typo、リンク修正、既存文言の明確化、形式調整など、one-way door がないと調査で 確認できた軽微な Issue はスキップする。Issue を変更せず、次の 1 行だけを返す。

軽微な変更で one-way door がないため grill-me は不要です。/issue-review-ready <issue_id> へ進んでください。

迷う場合は interview を続ける。

Step 3: decision tree を作る

チェックリストの判断軸を使い、調査後も残る未決事項を依存順に並べる。

  • one-way door を先に解決する。
  • two-way door は、人間の選好が必要な場合だけ質問する。その他は AI の仮定、根拠、 後段の検査先として記録候補にする。
  • 既に出典付きで決定済みの事項を聞き直さない。ただし、新しい情報との矛盾は質問する。
  • 固定質問リストを機械的に読み上げず、回答によって後続 branch を更新する。

Read the full file on GitHub · 151 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 10d ago First seen · 151 lines · 95 tokens per session scan A 0477dfa0c2f0

Subscribe to this mod's changes

grill-me is a skill published in the GitHub repository apokamo/kaji (12 stars, last pushed 3d ago), licensed Apache-2.0. It adds 95 tokens to every session and 2,121 once invoked, about $0.0005 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.