China Audit Xls logo

China Audit Xls

Community
jwangkun
china-audit-xls

Audit financial models and Excel workbooks for A-share analysis. Adapts the original audit-xls skill for Chinese financial modeling standards, CAS conventions, and A-share-specific checks. Triggers on "A股模型审计", "财务模型核查", "audit model China", "audit xlsx", "模型QC", or "check model [company]".

Overview

Publisherjwangkun
Repositoryclaude-for-financial-services-cn
Skill namechina-audit-xls
Stars
834
Forks
132
Bundled files
Instructions only
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by jwangkun on GitHub. Read the source before you install it.

Installation

Install the China Audit Xls AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/jwangkun/claude-for-financial-services-cn/tree/main/agent-plugins/china-earnings-reviewer/skills/china-audit-xls
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/jwangkun/claude-for-financial-services-cn.git /tmp/claude-for-financial-services-cn
mkdir -p .claude/skills
cp -r /tmp/claude-for-financial-services-cn/agent-plugins/china-earnings-reviewer/skills/china-audit-xls .claude/skills/china-audit-xls
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable China Audit Xls in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads China Audit Xls on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so China Audit Xls is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

china-audit-xls

Purpose

Audit A股财务模型 — comprehensive quality checks for Chinese equity financial models.

Data Sources

Tier 0 — 万得 Wind(最全面付费数据)

  • 覆盖:A股/港美股/基金/指数/债券/宏观/研报/分析(44个工具)
  • MCP 服务:wind-mcp(需 WIND_API_KEY 密钥,以 ak_ 开头)
  • 优势:全市场覆盖面最广、数据最全面、包含研报和量化分析
  • 密钥申请:https://aifinmarket.wind.com.cn/#/home

Tier 1 — 同花顺 iFind(付费精确数据)/ AkShare MCP(Tier-2 免费备选)

python
get_financials(ticker, "income")     → Actuals for cross-check
get_financials(ticker, "balance")    → BS for cross-check
get_financials(ticker, "cashflow")   → CF for cross-check
get_quote(ticker)                    → Market data

Secondary Sources

  • 巨潮 — source financials
  • 审计报告 — reference data

Workflow

Step 1: Structural Audit

Model structure checklist:

AreaCheckPass Criteria
Cover pageCompany, date, versionPresent and accurate
AssumptionsAll key inputs centralizedNo hardcoded values in calc
HistoricalsAll periods populated3-5 years
ForecastExplicit forecast period3-5 years
ValuationDCF, comps, football fieldAll methods present
ChecksSum, balance, cross-refAll checks green
DocumentationMethodology notesClear and complete

Step 2: Formula Audit

Formula checks:

CheckDescriptionPass Criteria
No hardcodesAll inputs in assumptions✓
Consistent formulasSame formula across periods✓
No circularity (unless intended)Circular refs flagged✓
Error handlingIFERROR used where needed✓
Named rangesKey cells named✓
Sheet referencesCross-sheet refs work✓
Broken linksNo external links or all work✓

Common hardcodes to find:

  • Growth rates embedded in formulas
  • Multiples typed into cells
  • Shares outstanding hardcoded
  • Tax rates not in assumptions

Step 3: Historicals Cross-Check

Cross-check against source data:

Line ItemModeliFind / AkShare / 巨潮DifferenceExplanation
营业收入
营业成本
归母净利润
总资产
净资产
经营现金流

Cross-check tolerance:

  • Revenue, profit: ±2%
  • Balance sheet: ±1%
  • Cash flow: ±3% (timing differences)

Step 4: CAS Compliance Check

CAS-specific checks:

CheckDescription
Revenue net of VAT营业收入 not gross
R&D expensedNot capitalized (unless criteria met)
Tax calculation25% / 15% rate applied correctly
Minority interestSeparated from parent equity
Government grantsCorrect classification
合同负债Recognized appropriately
信用减值损失CAS 22 model applied

Step 5: Balance Sheet Integrity

BS checks:

CheckFormulaPass
BS balancesAssets = L + E✓
Cash flow tiesEnding cash = Beginning + Net CF✓
Debt scheduleShort + Long = Total debt✓
Share countShares × Price = Market cap✓
Minority interestCorrect % applied✓

Step 6: Forecast Logic Check

Forecast quality:

CheckDescriptionPass
Growth reasonableWithin industry range✓
Margins stableNo unexplained jumps✓
Working capitalTrends with revenue✓
CapExConsistent with depreciation✓
Debt serviceCan be serviced from FCF✓
DividendConsistent with policy✓

Growth rate benchmarks:

MetricConservativeBaseOptimistic
Revenue growth5-10%10-20%20-30%
Margin expansion0 ppts0-2 ppts2-5 ppts
Tax rate25%25%15% (if 高新)

Step 7: Valuation Check

Valuation audit:

MethodCheckPass
DCFWACC reasonable (6-10%)✓
DCFTerminal growth < GDP growth✓
DCFFCF positive in base case✓
CompsPeer group appropriate✓
CompsMultiples reasonable✓
Football fieldMin/median/max consistent✓

WACC components (China):

ComponentTypical Range
Risk-free rate (CGB 10Y)2.0-3.0%
Equity risk premium6-8%
Cost of debt3-6%
Tax rate25% (15% for 高新)

Step 8: Sensitivity & Scenario Check

Sensitivity audit:

CheckDescription
Sensitivity tablesKey variables tested
ScenariosBase / Bull / Bear
Tornado chartKey drivers identified
Monte Carlo (if used)Assumptions reasonable

Step 9: Common A-share Model Errors

Error checklist:

ErrorDetectionFix
收入用含税VAT not removedDivide by 1.13
税率错误Wrong rate appliedVerify 高新 status
单位错误千元 vs 万元Standardize
季度加总错误Q+Q ≠ AnnualCheck sum
增长率计算Wrong base periodVerify formula
少数股东损益MissingAdd if applicable
折旧年限WrongCheck FA notes

Step 10: Audit Report

Standard audit output:

【模型审计报告】[Company] [Date]

一、结构检查
   [Structure assessment]

二、公式检查
   [Formula audit results]

三、历史数据核对
   [Cross-check results]

四、CAS合规检查
   [CAS compliance]

五、预测逻辑检查
   [Forecast quality]

六、估值检查
   [Valuation audit]

七、发现的问题
   [List of issues with severity]

八、建议修复
   [Recommended fixes]

总体评价: [Pass / Pass with conditions / Fail]

Quality Checks

Before completing:

  • All structural checks passed
  • No hardcodes found
  • Historicals cross-checked
  • CAS compliant
  • BS/CF integrity verified
  • Forecast logic sound
  • Valuation reasonable
  • All issues documented

Data Source Mode Switch: Set env var IFIND_DATA_SOURCE_MODE to control data source preference.

  • ifind-only (strict): Use iFind only, error if unavailable
  • ifind-fallback (default): iFind preferred, fallback to AkShare
  • akshare-only: Skip iFind, use AkShare only
  • wind-only: Wind only, error if unavailable
  • wind-fallback: Wind first, fallback to iFind → AkShare

Frequently asked questions

What does the China Audit Xls AI skill do?

Audit financial models and Excel workbooks for A-share analysis. Adapts the original audit-xls skill for Chinese financial modeling standards, CAS conventions, and A-share-specific checks. Triggers on "A股模型审计", "财务模型核查", "audit model China", "audit xlsx", "模型QC", or "check model [company]".

Why use China Audit Xls on TypingMind?

Because you install it once and use it with any model. China Audit Xls is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install China Audit Xls in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/jwangkun/claude-for-financial-services-cn/tree/main/agent-plugins/china-earnings-reviewer/skills/china-audit-xls. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use China Audit Xls?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with China Audit Xls?

As many as you like. As long as a model supports skills, you can use China Audit Xls with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the China Audit Xls AI skill free?

Yes. It is published on GitHub by jwangkun under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇