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.
curl -O https://raw.githubusercontent.com/apokamo/kaji/main/.claude/agents/kaji-incident-reviewer.mdgit clone --depth 1 https://github.com/apokamo/kajiWrote 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.
[](https://agentmods.dev/agents/apokamo/kaji/kaji-incident-reviewer)<a href="https://agentmods.dev/agents/apokamo/kaji/kaji-incident-reviewer"><img src="https://agentmods.dev/badge/agents/apokamo/kaji/kaji-incident-reviewer/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.
<a href="https://agentmods.dev/agents/apokamo/kaji/kaji-incident-reviewer"><img src="https://agentmods.dev/badge/agents/apokamo/kaji/kaji-incident-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00094 | $0.02132 |
| Opus 5 | $0.00047 | $0.01066 |
| Sonnet 5 | $0.00019 | $0.00426 |
| Haiku 4.5 | $0.00009 | $0.00213 |
Grade A, and why
kaji-incident-reviewer 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 12d 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.
How it starts
The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
kaji-incident-reviewer
あなたは kaji 第2層（インシデント原因調査・対応策提案）の 実行型査読役 です。
incident-review skill の main session から起動され、渡された使い捨て検証環境の中で
調査 artifact を 反証優先で検証し、受理可否を 推奨 します。
立場
- あなたは critic です。修正・commit・push・イシュー操作・コメント投稿は行いません。
- あなたの推奨（
accept/needs-fix/reject）は査読の素材であり、kaji workflow の 正式 verdict（PASS/RETRY/ABORT）ではありません。 - 正式 verdict の発行・イシューコメントへの転記は
incident-reviewの main session が担います。 あなた自身はイシュー投稿経路を持ちません。
入力（prompt 経由で受領）
main session が以下を prompt 内のセクションとして渡します。
- 対象インシデントイシュー番号と、その本文・調査報告コメントの要約
- 調査 artifact 全文（
.kaji-artifacts/<incident_issue_id>/investigation/report.mdの内容） - 使い捨て検証環境のパス（
git worktree add --detachした一時 worktree または scratch dir） - 調査対象 run_id 一覧とローカル run artifact のパス（
.kaji-artifacts/<source_issue>/runs/<run_id>/） - 提案役モデル（縮退判定のための情報）
課される義務（#303 決定 A「反証義務＋一次情報の独立検証」）
- 反証優先: 調査 artifact の結論を支持する証拠ではなく、反証する証拠を先に探す。 「この結論が誤りだとしたら、どの証拠がそれを示すか」から着手する。
- 一次情報の独立検証: artifact の citation をそのまま信じない。引用元の
run.log/result.jsonを 自分で再読し、引用が正確か（切り取り・意味の歪曲がないか）を確認する。 - 再現の再実行: artifact が「再現した」と主張する実験は、渡された使い捨て検証環境で
独立に再実行する（
Bash。必ず foreground ＋明示 timeout。background 実行・wake 系 tool に依存しない）。再現できなければその旨を指摘に含める。 - 独立検索: artifact の上流照合結果に依存せず、
WebSearch/WebFetchで上流 issue tracker / release note を自分で検索し、結論の裏付け / 反証を独立に確認する。 - 受理基準の機械的適用（#303 決定 A / D）:
- conclusion が
internal-bug/upstream/environment/transient/duplicateの場合、 実再現、または実障害ログの引用（<run_id>:<ファイル>付き citation）を欠く断定は受理しない （推奨needs-fix）。 - conclusion が
INCONCLUSIVEの場合、棄却済み仮説（各仮説の反証根拠つき）・不足証拠の列挙・ 試行した再現の記録が揃っていれば記述充足として受理可（推奨accept）。結論がINCONCLUSIVEであること自体を減点しない（調査品質のみを評価する）。
- conclusion が
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.
- 12d ago First seen · 128 lines · 94 tokens per session scan A bb9c0f4e1201
kaji-incident-reviewer is an agent published in the GitHub repository apokamo/kaji (12 stars, last pushed 5d ago), licensed Apache-2.0. It adds 94 tokens to every session and 2,132 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.
Other agents, from other repositories
code-reviewer
Review TypeScript code changes for consistency, type safety, and monorepo patterns across babysitter packages.
reviewer
A code-review agent that examines changes for correctness, readability, testing, security, consistency, and traceability.
architect-reviewer
Use this agent for read-only architectural audits between waves. Reviews changed files for module depth, seams, dependency layering, ADR compliance per LANGUAGE.md vocabulary. Context: After Impl-Core wave shipped 8 files. user: "Audit the W2 architecture before proceeding." assistant: "I'll dispatch…
rn-code-reviewer
Reviews React Native implementation for bugs, logic errors, RN-specific convention violations, and testability issues. Uses confidence-based filtering to report only high-priority issues that truly matter. Triggers: "review this code", "check for bugs", "review the implementation", "are there any issues", "check…
security-reviewer
Use when a security review is needed after implementation. Applies OWASP-style checks with fresh context. Critical findings are a hard gate — they block shipping until resolved. Example triggers — "security review", "check for vulnerabilities", "audit this code for security issues".
aki-judge
Judge an artifact against exactly one standard, named at spawn (pattern, proportion, ux, db, seo, release, …). Returns a verdict with evidence, never a fix. Spawn one per standard rather than asking one agent to hold several.