跳到正文
cccskills
免费在 GitHub 公开

audit-doc-checker

Review audit-related documents (control design documents, bottleneck analyses, requirements definitions, etc.) for quality, scoring them 0-100 with a severity-rated findings list. Use when reviewing audit documents, checking control design quality, or verifying cross-document consistency. Supports documents governed by US GAAP, IFRS, or J-GAAP.

中文简介正在准备中。当前显示原文说明。

实测评分 未实测

查看安装方法

包含的文件(4)

  • SKILL.md5.7 KB
  • assets/review_output_template.md2.5 KB
  • references/check_rules.md16.4 KB
  • references/scoring_model.md8.0 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Audit Document Quality Checker

Overview

This skill reviews audit-related documents against 12 quality check categories and produces a structured quality score (0-100) with a detailed findings list. Each finding includes severity (High/Medium/Low), location in the document, description, and recommended fix.

When to Use

  • Reviewing a control design document before submission to external auditors
  • Checking the quality of a bottleneck analysis or risk assessment report
  • Verifying consistency across audit-related documentation (terminology, currency, accounting standards)
  • Validating that a document covers all required audit assertions (C/A/V/CO/E)
  • Ensuring open questions and TBD items are properly tracked
  • Pre-publication quality gate for any document that will be used in an audit context

Prerequisites

None. This is a knowledge-based skill that uses reference documents to guide the review.

Workflow

Step 1: Identify the Target Document

Read the document to be reviewed. Determine the document type:

Document TypeDescriptionPriority Categories
Control Design DocumentInternal control procedures, SoD analysis, KPIsCategories 5, 7, 8 (weighted higher)
Bottleneck AnalysisProcess bottleneck identification and risk assessmentCategories 3, 5, 6
Requirements DefinitionBusiness or system requirements for audit-related systemsCategories 3, 6, 9
Audit ReportFindings, recommendations, management responsesCategories 1, 9, 12
Process InventoryAs-Is or To-Be process documentationCategories 4, 9, 11

Step 2: Load Check Rules

Load references/check_rules.md to obtain the 12 check categories and their detailed rules.

Step 3: Analyze Against 12 Categories

Systematically review the document against each of the 12 categories:

  1. Terminology Consistency (10 pts)
  2. Currency/Unit Consistency (10 pts)
  3. Accounting Standards Alignment (15 pts)
  4. Section Numbering/Cross-References (5 pts)
  5. Control Logic Consistency (15 pts)
  6. Materiality Criteria Application (10 pts)
  7. Assertion Coverage (10 pts)
  8. SoD Analysis Presence (5 pts)
  9. Open Items Management (10 pts)
  10. Preventive/Detective Classification (5 pts)
  11. Success Criteria Definition (5 pts)
  12. Metadata Freshness (5 pts)

For each category, record findings with severity:

  • High: Audit risk — could lead to material misstatement or regulatory non-compliance
  • Medium: Quality gap — reduces document reliability but not immediately dangerous
  • Low: Cosmetic — formatting, minor inconsistencies that don't affect substance

Step 4: Calculate Score

Load references/scoring_model.md and apply the scoring model:

  1. Determine the document type (Step 1) and look up the multiplier table in references/scoring_model.md Section 4
  2. For each category, calculate weighted deductions: Weighted_Deduction = Σ (Severity_Points × Multiplier) where Severity_Points are High=5, Medium=3, Low=1 and Multiplier is ×1.5 (priority), ×1.0 (standard), or ×0.5 (lower-priority)
  3. Calculate category score: Category_Score = max(Base_Points - Weighted_Deduction, 0)
  4. Sum all 12 category scores for the total (0-100)

Step 5: Generate Output

Use assets/review_output_template.md to format the final output:

  1. Score summary table
  2. Category-by-category score breakdown
  3. Findings list (sorted by severity, then category)
  4. Overall assessment paragraph

Output Format

The output follows the template in assets/review_output_template.md and includes:

  • Total Score: 0-100 with quality tier label
  • Score Tiers: 90+ (High Quality), 70-89 (Improvement Recommended), 50-69 (Revision Required), <50 (Critical Risk)
  • Category Breakdown: Per-category score and finding count
  • Findings Table: Severity, category, location, description, recommended fix
  • Overall Assessment: 1-2 paragraph summary with prioritized action items

Quality Tier Definitions

TierScore RangeMeaningRecommended Action
High Quality90-100Ready for audit submissionMinor polish only
Improvement Recommended70-89Fundamentally sound but has gapsAddress Medium/High findings
Revision Required50-69Significant gaps that undermine reliabilityMajor revision before use
Critical Risk0-49Document is not audit-readyRebuild with proper framework

Resources

TypeFilePurposeWhen to Load
Referencereferences/check_rules.md12 check categories with detailed rules, severity criteria, and examplesStep 3: Load before analyzing the document
Referencereferences/scoring_model.mdScoring calculation: base points, deduction rules, document-type multipliers, tier definitionsStep 4: Load before calculating score
Assetassets/review_output_template.mdOutput template with placeholder variables for structured review resultsStep 5: Use as output format

Integration with audit-control-designer

This skill can be used to review documents generated by the audit-control-designer skill. The recommended workflow is:

  1. Generate control design with audit-control-designer
  2. Review the generated document with audit-doc-checker
  3. Address findings and iterate until score reaches 70+

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Track and update action item status from natural language updates like 'Seanのメールには返信しておいた' or 'Lu対応予定'. Integrates with daily-comms-ops workflow. Use when managing action items across communication channels.

中文简介正在准备中。当前显示原文说明。

takusaotome/claude-skills-library102026年10月5日 更新

AI活用提案の専門コンサルタントスキル。LLM・AIエージェントの活用事例知識を基に、業界別、部門別、シーン別に最適なAI導入提案を行います。システム化、業務改善、効率化を得意とし、具体的な活用事例、期待効果、導入ステップを含む実践的な提案を提供します。Use when proposing AI/LLM adoption strategies for specific industries, departments, or business scenarios.

中文简介正在准备中。当前显示原文说明。

takusaotome/claude-skills-library102026年10月5日 更新

Generate AI-powered BPO service proposals for Japanese companies in the US market. Use when creating outsourcing proposals with AI service menus, ROI estimation, implementation roadmaps, and bilingual (JA/EN) proposal documents for 在米日系企業向けAI実装.

中文简介正在准备中。当前显示原文说明。

takusaotome/claude-skills-library102026年10月5日 更新

AI(LLM)が生成した日本語テキストの「AI臭」を検出・診断し、人間らしい文章にリライトするスキル。 Use when: 「AIっぽい文章を直して」「人間らしくリライトして」「AI臭を消して」「この文章をもっと自然にして」 「テキストのAI感を減らして」「機械っぽさを取りたい」 "make this sound more human", "remove AI tone", "humanize this text", "detect if this is AI-generated". 6つのAI特有パターン(視覚的マーカー、単調なリズム、マニュアル的構成、非コミット姿勢、抽象語の濫用、 定型メタファー)を正規表現ベースで検出し、0-100のAI臭スコアを算出。3つの人間化技法 (バランスを崩す・客観を崩す・論理を崩す)でリライトを実行する。 Note: 検出スクリプトは日本語テキスト専用。英語テキストの場合はClaude自身がreferences/を参照して分析・リライトする。

中文简介正在准备中。当前显示原文说明。

takusaotome/claude-skills-library102026年10月5日 更新

Generate audit-ready internal control design documents from As-Is business process inventories. Produces control IDs, assertion mappings, procedures, SoD analysis, KPIs, materiality thresholds, and implementation roadmaps. Use when building internal controls for new audit engagements, SOX/J-SOX compliance, or process improvement initiatives.

中文简介正在准备中。当前显示原文说明。

takusaotome/claude-skills-library102026年10月5日 更新

AWS CLIの専門家スキル。EC2, S3, IAM, Lambda, RDS, ECS等の主要サービスのCLI操作を支援。IAMポリシー設計、クロスアカウント運用、セキュリティ設定、コスト最適化のベストプラクティスを提供。Use when working with AWS services via CLI, designing IAM policies, managing infrastructure, or automating cloud operations.

中文简介正在准备中。当前显示原文说明。

takusaotome/claude-skills-library102026年10月5日 更新

查看 takusaotome 的全部技能

报告此技能的问题