AI Agent Hub
Back to skills
Financial Data Visualization Analysis Assistant icon

Financial Data Visualization Analysis Assistant

Data Analysis Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Install @user_7ea4583e/financial-edu-david into your AI assistant using the guide at https://skillhub.cn/install/skillhub.md.

About this skill

The Problem Solved

When processing historical financial data for A-share listed companies, engineers often face specific pain points: annual report PDFs from CNINFO have complex layouts where account names are frequently split across lines and mixed with footnote references; manually consolidating 3 to 5 years of consolidated financial statements is time-consuming and prone to unit (CNY to 100M CNY) or mapping errors; furthermore, the lack of automated reconciliation with official public data (like East Money F10) undermines the reliability of the analysis.

Core Capabilities and Workflow

The skill decomposes financial analysis into a strict five-stage pipeline:
- Data Collection and Parsing: Automatically fetches annual report PDFs. It uses pdfplumber to detect page numbers, applies a pending_subject state machine to handle cross-line account names, cleans footnote references via regex, and outputs the three statements into xlsx files.
- Metric Calculation: Extracts 30+ core accounts from the structured data and standardizes units to 100M CNY. It automatically calculates solvency, operational, and profitability metrics, constructing a DuPont analysis tree based on ROE = Net Profit Margin × Total Asset Turnover × Equity Multiplier.
- Visualization and Validation: Generates a single-file HTML dashboard with 8 embedded ECharts trend charts and a pure HTML+CSS DuPont tree. Before delivery, it cross-references core accounts using the East Money F10 API, correcting the underlying JSON with F10 data if discrepancies are found.

Boundaries and Considerations

  • The workflow heavily relies on external network APIs (CNINFO, East Money). If upstream APIs change, request parameters in cninfo_api.md must be manually adjusted.
  • PDF parsing has inherent noise; income statement accounts allow a ±5% error margin, but balance sheet and cash flow metrics must match 100%, relying on the F10 API as a fallback for anomalies.
  • Requires a Python 3.8+ runtime environment and write permissions for the working directory.

Use Cases

  • An equity researcher parses 4 years of A-share annual report PDFs into three statements and calculates ROE trends.
  • A financial analyst generates an HTML report with cash flow, expense ratios, and a DuPont tree for a specified stock.
  • A data engineer extracts consolidated financial statements from CNINFO PDFs into xlsx, handling split account names and footnotes.
  • A risk reviewer reconciles extracted metrics with East Money F10 and corrects asset, liability, and cash flow discrepancies.

Best For

  • Sell-side analysts covering A-share research who need multi-year statements, metrics, and visualization working papers quickly.
  • Financial advisors performing public-company due diligence who need annual report PDFs converted into reconciled metric tables and dashboards.
  • Data engineers building finance automation scripts who need reliable CNINFO PDF extraction into xlsx and JSON datasets.
  • Risk reviewers in credit or investment teams who need cash flow, leverage, and ROE reconciled against East Money F10.