Financial Fraud Index Analysis
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md to install @user_4f8090ec/financial-fraud-index.
About this skill
Problem
Fraud-risk review of financial statements is slowed by three issues: long reports, scattered evidence, and overconfident conclusions. Annual reports, audit reports, and statement notes often require separating confirmed anomalies from weak signals and missing data rather than relying on summaries alone.
How It Works
The skill follows an evidence chain. It confirms source type, company, period, and input format; extracts key fields, notes, audit opinion, and governance disclosures; then classifies findings into confirmed anomalies, weak signals, and missing data. Confirmed anomalies must cite evidence, preferably in the form 原文摘录（第X页）; calculated ratios must show source values and calculation basis. The output follows a fixed order: report summary, confirmed anomalies, weak signals, missing data, evidence excerpts, and review guidance.
Boundaries
It is suited to analyzing supplied reports for risk indicators, not unsourced investment advice, broad market commentary, or valuation modeling disconnected from report evidence. If cash-flow support, sales receipts, note excerpts, or page anchors are missing, conclusions should be downgraded to pending review rather than stated as definitive ratings.
Use Cases
- Review annual or audit reports to organize revenue, profit, and cash-flow contradictions into page-referenced red flags.
- Compare companies or periods to separate confirmed anomalies, weak signals, and missing fields without overstating conclusions.
- Extract notes, audit opinions, and governance disclosures from PDF reports or extracted text into a review checklist.
- Flag missing cash-flow, sales receipts, or provision notes as pending review instead of giving a definitive rating.
Best For
- Risk analysts reviewing annual reports who need red flags tied to page numbers, fields, and notes.
- Credit or investment due-diligence analysts who need reviewable risk signals from statements and audit opinions.
- Audit assistants or compliance specialists who need to split long reports into anomalies, weak signals, and gaps.
- Due-diligence leads who need consistent summaries, evidence excerpts, and next-step review guidance.
Related Skills
A local A-share quant workspace for quotes, k-lines, conditional screening, MA/RSI backtesting, simulated trading, and risk control via REST APIs.
Systematically identify and evaluate the economic moats of listed companies using the Tang Shu Fang investment methodology for long-term investment analysis.
A skill that converts natural language questions into A-stock data queries and returns verifiable structured analysis conclusions, covering multi-dimensional analysis of market trends, fundamentals, and news.
An AI financial copilot by Wind, integrating financial databases and multimodal analysis to provide end-to-end support across investment research, asset allocation, risk control, quant, and report generation.