Ipa Security Check

What it does

When your agent uses it

• Tasks that involve SQL

Example prompts

• “/ipa-security-check”

Requirements

• Python 3

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

1. 1lib/scope_resolver.md を読み、引数を解釈して対象ファイル一覧と言語マッピングを作る
2. 2lib/shard_planner.md を読み、カテゴリごとにファイル数を数え、閾値超過時は N 分割する
3. 3agents/ 配下 の検査系サブエージェント (14 体) を メインから並列起動 する (公式制約によりサブエージェントから二次サブエージェントは起動できないため、分割はメイン側で行う)
4. 4各サブエージェントは findings[] を含む JSON を返す。メインは findings のみ集約してコード本文は文脈に保持せず、.tmp/ipa-security-check/findings_raw.json に書き出す
5. 5Phase 5 (偽陽性レビュー)
6. 6Phase 6 + 出力 (Step 7): scripts/render_report.py を Bash で実行する。スクリプトが内部で以下を行う
7. 7デフォルトの保存先は ./security-reports/ipa-security-report-YYYY-MM-DD-NN.md と ./security-reports/ipa-security-report-YYYY-MM-DD-NN.sarif (--output…

What it can do on your machine

Read from SKILL.md and the folder at commit fa5aaff. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 1 file in scripts/, which the agent can run.

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Hosts in commands or code, which the agent is likely to contact:

  • ipa.go.jp

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Ipa Security Check loads about 1.8k tokens when it runs. Until then it costs about 46 tokens; SKILL.md has 442 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~46

When it runs · the whole SKILL.md, loaded when a task matches

~1.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from classmethod/tsumiki at commit fa5aaff, republished under its MIT licence (© classmethod). 442 words, ~1,781 tokens.

name

ipa-security-check

description

IPA「安全なウェブサイトの作り方 改訂第7版」「安全なSQLの呼び出し方」「ウェブ健康診断仕様」「セキュリティ実装チェックリスト」「安全なウェブサイトの運用管理に向けての20ヶ条」に基づき、ソースコードを静的に検査して脆弱性候補を検出する。発見した問題には IPA 原典の出典 (文書名・章・ページ・URL) を必ず付与する。

// ipa-skip: IPA-SWS-1-SQLI-001  reason: 内部固定値を埋め込んでいるため

{
  "agent": "sql-injection",
  "files_scanned": 142,
  "findings": [
    {
      "rule_id": "IPA-SWS-1-SQLI-001",
      "severity": "critical",
      "category": "sql_injection",
      "file": "src/users.php",
      "line": 45,
      "column": 12,
      "code_snippet": "$sql = \"SELECT * FROM users WHERE id = \" . $_GET['id'];",
      "message": "...",
      "ipa": {
        "document": "安全なウェブサイトの作り方 改訂第7版",
        "section": "1.1 SQLインジェクション",
        "page": "6-12",
        "url": "https://www.ipa.go.jp/security/vuln/websecurity/about.html"
      },
      "remediation_type": "根本的解決",
      "remediation": "プレースホルダによる SQL 文の組み立て",
      "cwe": "CWE-89",
      "fix_example": "..."
    }
  ],
  "errors": []
}

.claude/skills/ipa-security-check/
├── SKILL.md                ← 本ファイル (Claude が最初に読む)
├── commands/
│   └── ipa-security-check.md
├── agents/                 ← 15 体のサブエージェント定義
│   ├── 01〜11, safe-sql-details, web-health-check, operation-checklist (検査 14 体)
│   └── 15-false-positive-review.md (偽陽性レビュー)
├── knowledge/              ← IPA 原文ベースの知識 (Skill 単独配布で完結)
├── rules/                  ← YAML 検出シグネチャ
├── templates/              ← Markdown / SARIF テンプレート
├── scripts/                ← Bash 経由で呼ぶ Python 実装
│   ├── snippet_hash.py     ← snippet_hash 計算 (Phase 5 入力準備)
│   └── render_report.py    ← verdict/triage マージ + Markdown/SARIF 出力 (Step 7)
└── lib/
    ├── orchestrator.md
    ├── scope_resolver.md
    ├── shard_planner.md
    ├── triage_state.md     ← ステータス保持・引き継ぎ仕様
    └── output_formatter.md

© classmethod, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 58 other files (scripts) in skills/ipa-security-check of classmethod/tsumiki.

• SKILL.md
• agents/01-sql-injection.md
• agents/02-os-command-injection.md
• agents/03-directory-traversal.md
• agents/04-session-management.md
• agents/05-xss.md
• agents/06-csrf.md
• agents/07-http-header-injection.md
• agents/08-mail-header-injection.md
• agents/09-clickjacking.md
• agents/10-buffer-overflow.md
• agents/11-access-control.md
• agents/15-false-positive-review.md
• agents/operation-checklist.md
• agents/safe-sql-details.md
• agents/web-health-check.md
• commands/ipa-security-check.md
• knowledge/01_sql_injection.md
• … and 41 more

Open the folder on GitHubat commit fa5aaff